Kernel Regression in the Presence of Correlated Errors Kernel Regression in the Presence in nonparametric regression is difficult in the presence of correlated errors. There exist a wide variety vector machines for regression. Keywords: nonparametric regression, correlated errors, bandwidth choice
SUBSPACE REGRESSION IN REPRODUCING KERNEL HILBERT SPACE
SUBSPACE REGRESSION IN REPRODUCING KERNEL HILBERT SPACE L. Hoegaerts #3; J.A.K. Suykens #3; J: We focus on three methods for finding a suitable subspace for regression in a reproducing kernel correlation analysis and we demonstrate how this fits within a more general context of subspace regression
Multivariate wavelet kernel regression method Samir Touzani, Daniel Busbya
Paris-Sud XI, Université de
Multivariate wavelet kernel regression method Samir Touzani, Daniel Busbya a IFP Energies nouvelles multivariate nonparametric regression method, in the framework of wavelet decomposition. We call this method the wavelet kernel ANOVA (WK-ANOVA), which is a wavelet based reproducing kernel Hilbert space (RKHS) method
Kernel Density Based Linear Regression Estimate and Zhibiao Zhao
Zhao, Zhibiao
Kernel Density Based Linear Regression Estimate Weixin Yao and Zhibiao Zhao Abstract For linear regression models with non-normally distributed errors, the least squares estimate (LSE) will lose some words: EM algorithm, Kernel density estimate, Least squares estimate, Linear regression, Maximum
Quantile Regression with 1-regularization and Gaussian Kernels
Quantile Regression with 1-regularization and Gaussian Kernels Lei Shi1,2 , Xiaolin Huang1 , Zheng Sciences, Fudan University Shanghai 200433, P. R. China Abstract The quantile regression problem of 1-quantile regression with Gaussian kernels is almost the same as that of the RKHS-based learning
Using the Equivalent Kernel to Understand Gaussian Process Regression
Sollich, Peter
Using the Equivalent Kernel to Understand Gaussian Process Regression Peter Sollich Dept.k.i.williams@ed.ac.uk Abstract The equivalent kernel [1] is a way of understanding how Gaussian pro cess regression works processes. Consider the supervised regression problem for a dataset D with entries (x i ; y i ) for i = 1
Using the Equivalent Kernel to Understand Gaussian Process Regression
Sollich, Peter
Using the Equivalent Kernel to Understand Gaussian Process Regression Peter Sollich Dept.k.i.williams@ed.ac.uk Abstract The equivalent kernel [1] is a way of understanding how Gaussian pro- cess regression works processes. Consider the supervised regression problem for a dataset D with entries (xi, yi) for i = 1
FAST SPEAKER ADAPTION VIA MAXIMUM PENALIZED LIKELIHOOD KERNEL REGRESSION
Tsang Wai Hung "Ivor"
of MLLR using non- linear regression. Specifically, kernel regression is applied with appropriate of Science and Technology Clear Water Bay, Hong Kong ABSTRACT Maximum likelihood linear regression (MLLR) has], and transformation-based methods, most notably, maximum likelihood linear regression (MLLR) adap- tation [3]. However
Kernel Regression with Correlated Errors K. De Brabanter
Kernel Regression with Correlated Errors K. De Brabanter , J. De Brabanter , , J.A.K. Suykens B: It is a well-known problem that obtaining a correct bandwidth in nonparametric regression is difficult support vector machines for regression. Keywords: nonparametric regression, correlated errors, short
Kernel Machine Based Feature Extraction Algorithms for Regression Problems
Szepesvari, Csaba
Kernel Machine Based Feature Extraction Algorithms for Regression Problems Csaba Szepesv´ari 1 feature extraction algorithms in a regression settings. The first method is derived based. However, here it is shown that the orthogonalization principle employed by the original MMDA algorithm can
Feature Preserving Point Set Surfaces based on Non-Linear Kernel Regression
Kazhdan, Michael
Feature Preserving Point Set Surfaces based on Non-Linear Kernel Regression A. C. Öztireli, G IMLS in terms of Local Kernel Regression (LKR) · Borrowing ideas from robust statistics · Advantages to Implement · Competitive performance #12;Local Kernel Regression · Taylor expansion around the evaluation
Kernel Regression For Determining Photometric Redshifts From Sloan Broadband Photometry
D. Wang; Y. X. Zhang; C. Liu; Y. H. Zhao
2007-06-20T23:59:59.000Z
We present a new approach, kernel regression, to determine photometric redshifts for 399,929 galaxies in the Fifth Data Release of the Sloan Digital Sky Survey (SDSS). In our case, kernel regression is a weighted average of spectral redshifts of the neighbors for a query point, where higher weights are associated with points that are closer to the query point. One important design decision when using kernel regression is the choice of the bandwidth. We apply 10-fold cross-validation to choose the optimal bandwidth, which is obtained as the cross-validation error approaches the minimum. The experiments show that the optimal bandwidth is different for diverse input patterns, the least rms error of photometric redshift estimation arrives at 0.019 using color+eClass as the inputs, the less rms error amounts to 0.020 using ugriz+eClass as the inputs. Here eClass is a galaxy spectra type. Then the little rms scatter is 0.021 with color+r as the inputs.
Functional inverse regression and reproducing kernel Hilbert space
Ren, Haobo
2006-10-30T23:59:59.000Z
and Reproducing Kernel Hilbert Space. (August 2005) Haobo Ren, B.S., Peking University; M.S., Peking University Chair of Advisory Committee: Dr. Tailen Hsing The basic philosophy of Functional Data Analysis (FDA) is to think of the observed data functions... component analysis of ? . Duan and Li (1991) and Li (1997) presented more delicate results for analyzing single- index regression by SIR, Hsing and Carroll (1992) and Zhu and Ng (1995) derived the large sample properties of SIR based on ?-delta, Chen and Li...
Functional inverse regression and reproducing kernel Hilbert space
Ren, Haobo
2006-10-30T23:59:59.000Z
, . . ., betaprimepx}, for universalb element Rd. See Hall and Li (1993) for more detailed discussion of the condition. For a model satisfying this condition, the space span(?beta1, ?beta2, ?. . ., betap) contains the centered IR curve E(x|y) - E(x), where ? = Cov... quantile to estimate reference curves in clinical studies. For other IR approaches, Zhu and Fang (1996) proposed a kernel regression to estimate E(x|y) and also gave an asymptotic result. Gather, Hilker and Becker (2002) evaluated the sensitivity of SIR...
Kernel regression estimates of time delays between gravitationally lensed fluxes
Otaibi, Sultanah AL; Cuevas-Tello, Juan C; Mandel, Ilya; Raychaudhury, Somak
2015-01-01T23:59:59.000Z
Strongly lensed variable quasars can serve as precise cosmological probes, provided that time delays between the image fluxes can be accurately measured. A number of methods have been proposed to address this problem. In this paper, we explore in detail a new approach based on kernel regression estimates, which is able to estimate a single time delay given several datasets for the same quasar. We develop realistic artificial data sets in order to carry out controlled experiments to test of performance of this new approach. We also test our method on real data from strongly lensed quasar Q0957+561 and compare our estimates against existing results.
Parameter-insensitive kernel in extreme learning for non-linear support vector regression
Verleysen, Michel
Parameter-insensitive kernel in extreme learning for non-linear support vector regression Beno for regression which uses the e-sensitive loss and produces sparse models. However, non-linear SVRs are difficult.g. [24]). Used in conjunction with kernels, SVRs are powerful non-linear models for regression which
Representations of the LL BFKL Kernel and the Bootstrap
G. P. Vacca
2005-09-20T23:59:59.000Z
Different forms of the BFKL kernel both in coordinate and momentum representations may appear as a result of different gauge choices and/or inner scalar products of the color singlet states. We study a spectral representation of the BFKL kernel not defined on the Moebius space of functions but on a deformation of it, which provides the usual bootstrap property due to gluon reggeization. In this space the corresponding symmetry is made explicit introducing a deformed realization of the sl(2,C) algebra.
Kernel Regression with Order Preferences Xiaojin Zhu and Andrew B. Goldberg
Zhu, Xiaojin "Jerry"
such knowledge as positive correlation can be difficult in non-linear kernel regression, because of the non-linear, but the exact re- lation is highly non-linear and unknown. We can, however, easily create order preferencesKernel Regression with Order Preferences Xiaojin Zhu and Andrew B. Goldberg Department of Computer
Approximate l-fold cross-validation with Least Squares SVM and Kernel Ridge Regression
Edwards, Richard E [ORNL] [ORNL; Zhang, Hao [ORNL] [ORNL; Parker, Lynne Edwards [ORNL] [ORNL; New, Joshua Ryan [ORNL] [ORNL
2013-01-01T23:59:59.000Z
Kernel methods have difficulties scaling to large modern data sets. The scalability issues are based on computational and memory requirements for working with a large matrix. These requirements have been addressed over the years by using low-rank kernel approximations or by improving the solvers scalability. However, Least Squares Support VectorMachines (LS-SVM), a popular SVM variant, and Kernel Ridge Regression still have several scalability issues. In particular, the O(n^3) computational complexity for solving a single model, and the overall computational complexity associated with tuning hyperparameters are still major problems. We address these problems by introducing an O(n log n) approximate l-fold cross-validation method that uses a multi-level circulant matrix to approximate the kernel. In addition, we prove our algorithm s computational complexity and present empirical runtimes on data sets with approximately 1 million data points. We also validate our approximate method s effectiveness at selecting hyperparameters on real world and standard benchmark data sets. Lastly, we provide experimental results on using a multi-level circulant kernel approximation to solve LS-SVM problems with hyperparameters selected using our method.
Deformed Spectral Representation of the BFKL Kernel and the Bootstrap for Gluon Reggeization
J. bartels; L. N. Lipatov; M. Salvadore; G. P. Vacca
2005-06-23T23:59:59.000Z
We investigate the space of functions in which the BFKL kernel acts. For the amplitudes which describe the scattering of colorless projectiles it is convenient to define, in transverse coordinates, the Moebius space in which the solutions to the BFKL equation vanish as the coordinates of the two reggeized gluons coincide. However, in order to fulfill the bootstrap relation for the BFKL kernel it is necessary to modify the space of functions. We define and investigate a new space of functions and show explicitly that the bootstrap relation is valid for the corresponding spectral form of the kernel. We calculate the generators of the resulting deformed representation of the sl(2,C) algebra.
Jaillet, Patrick
Parallel Gaussian Process Regression for Big Data: Low-Rank Representation Meets Markov comes at a cost of poor scalability in the data size. To improve its scalability, this paper presents with larger data). As a result, our LMA method can trade off between the size of the support set and the or
The curse of dimension in nonparametric regression
Kpotufe, Samory
2010-01-01T23:59:59.000Z
3.2.1 Kernel regression . . . . . . . . . . . . . . . . . .3.2.2 k-NN regression . . . . . . . . . . . . . . . . . .1.1 Nonparametric regression . . . . . . . . . . . 1.1.1
Bayesian Kernel Shaping for Learning Control
Ting, Jo-Anne; Kalakrishnan, Mrinal; Vijayakumar, Sethu; Schaal, Stefan
2008-01-01T23:59:59.000Z
In kernel-based regression learning, optimizing each kernel individually is useful when the data density, curvature of regression surfaces (or decision boundaries) or magnitude of output noise varies spatially. Previous ...
Bayesian Analysis of Curves Shape Variation through Registration and Regression
Telesca, Donatello
2015-01-01T23:59:59.000Z
Wehrly (1986). Kernel regression estimation using repeatedadaptive Bayesian penalized regression splines (P-splines).self-modeling nonlinear regression. The Annals of Statistics
MULTIVARIATE NONPARAMETRIC REGRESSION AND VISUALIZATION
Klemelä, Jussi
not be available in electronic format. Library of Congress Cataloging-in-Publication Data: Klemel¨a, Jussi AND CLASSIFICATION 1 Overview of Regression and Classification 3 2 Linear Methods and Extensions 77 3 Kernel Methods Visualization xxi I.4 Literature xxiii PART I METHODS OF REGRESSION AND CLASSIFICATION 1 Overview of Regression
Adaptive wiener image restoration kernel
Yuan, Ding (Henderson, NV)
2007-06-05T23:59:59.000Z
A method and device for restoration of electro-optical image data using an adaptive Wiener filter begins with constructing imaging system Optical Transfer Function, and the Fourier Transformations of the noise and the image. A spatial representation of the imaged object is restored by spatial convolution of the image using a Wiener restoration kernel.
The Kernel Recursive Least Squares Algorithm Yaakov Engel
Meir, Ron
prediction and channel equalization. Keywords: on-line learning, kernel methods, non-linear regression Experiments 21 5.1 Non-Linear Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 5 the permission of the authors. #12;Abstract We present a non-linear kernel-based version of the Recursive Least
Jan de Leeuw
2011-01-01T23:59:59.000Z
values. And then we perform an ordinary monotone regression.nally mention that monotone regression can be generalized inMONOTONE REGRESSION JAN DE LEEUW Abstract. This is an entry
Unsupervised State-Space Modelling Using Reproducing Kernels
Tobar, Felipe; Djuri?, Petar M.; Mandic, Danilo P.
2015-06-22T23:59:59.000Z
gradient- based methods. These include ridge regression (RR) in the offline case, and least mean square (LMS) and recursive least squares (RLS) in online cases. These linear estimation algorithms are the basis of kernel adaptive filters. [ACCEPTED... in red. the observation signal yt using kernels and an LMS-based update rule; see [41]. Kernel State-Space Model (KSSM): The adaptive version of the proposed method, where the predictions are gener- ated by propagating the particles of the state according...
Linear Regression and Support Vector Regression
Shi, Qinfeng "Javen"
Linear Regression and Support Vector Regression Paul Paisitkriangkrai paulp@cs.adelaide.edu.au The University of Adelaide 18 August 2014 #12;Outlines · Regression overview · Linear regression · Support vector regression · Machine learning tools available #12;Regression Overview CLUSTERING CLASSIFICATION REGRESSION
Daume III, Hal
Robust RVM Regression Using Sparse Outlier Model Kaushik Mitra, Ashok Veeraraghavan* and Rama, vashok, rama}@umiacs.umd.edu Abstract Kernel regression techniques such as Relevance Vector Machine (RVM) regression, Support Vector Regression and Gaussian processes are widely used for solving many com- puter
"(Operating System)" linux "(kernel)"
kernel kernel shell (FIXME) shell (interface) (command line) shell ---(Command Interpreter line ""(word) (meta) command line () IFS shell #12;* (White Space) * (Tab) * (Enter) (command-name) * * (alias) * (function) * shell (built-in) * $PATH 3) echo echo echo command line echo --- command
Slicing Regression: Dimension Reduction via Inverse Regression
Ker-Chau Li
2011-01-01T23:59:59.000Z
Semiparametric Additive Regression," unpublished manuscript.N . , and L i , K . C. (in press), "Slicing Regression: aLink-Free Regression Method," The Annals of Statistics.
Nonlinear models Nonlinear Regression
Penny, Will
Nonlinear models Will Penny Nonlinear Regression Nonlinear Regression Priors Energies Posterior, UCL, March 2013 #12;Nonlinear models Will Penny Nonlinear Regression Nonlinear Regression Priors Example Sampling Metropolis-Hasting Proposal density References Nonlinear Regression We consider
Quantum Energy Regression using Scattering Transforms
Matthew Hirn; Nicolas Poilvert; Stephane Mallat
2015-02-06T23:59:59.000Z
We present a novel approach to the regression of quantum mechanical energies based on a scattering transform of an intermediate electron density representation. A scattering transform is a deep convolution network computed with a cascade of multiscale wavelet transforms. It possesses appropriate invariant and stability properties for quantum energy regression. This new framework removes fundamental limitations of Coulomb matrix based energy regressions, and numerical experiments give state-of-the-art accuracy over planar molecules.
Quantum Energy Regression using Scattering Transforms
Hirn, Matthew; Mallat, Stephane
2015-01-01T23:59:59.000Z
We present a novel approach to the regression of quantum mechanical energies based on a scattering transform of an intermediate electron density representation. A scattering transform is a deep convolution network computed with a cascade of multiscale wavelet transforms. It possesses appropriate invariant and stability properties for quantum energy regression. This new framework removes fundamental limitations of Coulomb matrix based energy regressions, and numerical experiments give state-of-the-art accuracy over planar molecules.
Wavelet Kernel Learning F. Yger and A. Rakotomamonjy1
Paris-Sud XI, Université de
Wavelet Kernel Learning F. Yger and A. Rakotomamonjy1 Universit´e de Rouen, LITIS EA 4108, 76800 from a wavelet representation. Our work aims at building features by selecting wavelet coefficients resulting from signal or image decomposition on a adapted wavelet basis. For this purpose, we jointly learn
West, Mike
of covariates to use in regression or generalized linear models is a ubiquitous problem. The Bayesian paradigm regression and binary re- gression with non-orthogonal design matrices in conjunction with independent "spike and kernel regression (Clyde and George 2004). The generalization of the Gaussian linear model to other
Jan de Leeuw
2011-01-01T23:59:59.000Z
Stone. Classi?cation and Regression Trees. Wadsworth, 1984.Projection Pursuit Regression. Journal of the AmericanK.C. Li. Sliced Inverse Regression for Dimension Reduction (
Gerber, Samuel
2013-01-01T23:59:59.000Z
1984). Classification and Regression Trees. Monterey, CA:Piecewise-polynomial regression trees. Statistica Sinica 4,BART: Bayesian additive regression trees. Ann. Appl. Stat.
Kernel Carpentry for Online Regression using Randomly Varying Coefficient Model
Edakunni, Narayanan U.; Schaal, Stefan; Vijayakumar, Sethu
2006-01-01T23:59:59.000Z
We present a Bayesian formulation of locally weighted learning (LWL) using the novel concept of a randomly varying coefficient model. Based on this
Machine Learning for Predictive Auto-Tuning with Boosted Regression Trees
Anderson, Charles H.
, kernel types, and platforms. 1. INTRODUCTION Due to power consumption and heat dissipation concerns for non-linear regression can be used to estimate timing models from data, capturing the best of both ap
KALMAN FILTERING REPRODUCING KERNEL HILBERT SPACES
Slatton, Clint
KALMAN FILTERING IN REPRODUCING KERNEL HILBERT SPACES Pingping Zhu #12;Outline · Introduction · Related Work · A Novel Extended Kernel Recursive Least Squares · Kernel Kalman Filter based on Conditional · Develop a Kalman filter in the Reproducing kernel Hilbert space (RKHS) Motivation · Kernel methods can
Orthogonal Forward Regression based on Directly Maximizing Model Generalization Capability
Chen, Sheng
for costly model evaluation. Index Terms -- orthogonal forward regression, structure identification, cross struc- ture construction process as a cost function in order to op- timize the model generalization introduces a construction algorithm for sparse kernel modelling using the leave-one-out test score also known
Regression problems for magnitudes
Castellaro, Silvia; Mulargia, Francesco; Kagan, Yan Y
2006-01-01T23:59:59.000Z
through general orthogonal regression with ? = 0.25 (? m2 =use and misuse of orthogonal regression in linear errors-in-A procedure for linear regression capable of less biased
Choquistic Regression: Generalizing Logistic Regression Using the Choquet Integral
Hüllermeier, Eyke
Choquistic Regression: Generalizing Logistic Regression Using the Choquet Integral Ali Fallah, called choquistic regression, which generalizes conventional logistic regression and takes advantage regression (2) Generalization to choquistic regression (3) First experimental results 2 #12;Logistic
Bruemmer, David J. (Idaho Falls, ID)
2009-11-17T23:59:59.000Z
A robot platform includes perceptors, locomotors, and a system controller. The system controller executes a robot intelligence kernel (RIK) that includes a multi-level architecture and a dynamic autonomy structure. The multi-level architecture includes a robot behavior level for defining robot behaviors, that incorporate robot attributes and a cognitive level for defining conduct modules that blend an adaptive interaction between predefined decision functions and the robot behaviors. The dynamic autonomy structure is configured for modifying a transaction capacity between an operator intervention and a robot initiative and may include multiple levels with at least a teleoperation mode configured to maximize the operator intervention and minimize the robot initiative and an autonomous mode configured to minimize the operator intervention and maximize the robot initiative. Within the RIK at least the cognitive level includes the dynamic autonomy structure.
Invariance in Kernel Methods by Haar-Integration Kernels
, regression, clustering, outlier-detection, feature- extraction etc. A powerful battery of algorithms
Aguiar, Pedro M. Q.
2009-01-01T23:59:59.000Z
inherent to the usual bag-of-words representations. In fact, approaches that map data to statistical/09 Nonextensive Information Theoretic Kernels on Measures Andr´e F. T. Martins AFM@CS.CMU.EDU Noah A. Smith, images, and other types of structured data. Some of these kernels are related to classic information
Local Regression and Likelihood
Masci, Frank
Local Regression and Likelihood Clive Loader Springer #12;#12;#12;#12;#12;Preface This book, and the associated software, have grown out of the author's work in the field of local regression over the past- ods and in particular regression, for example at the level of Draper and Smith (1981). The theoretical
Lega, Joceline
Introduction AI/Regression Technology/Search Simple Equations, Deep Understanding How Topics and Alison Cornell Simple Equations, Deep Meaning #12;Introduction AI/Regression Technology/Search Who are we helps out) Qiyam Tung and Alison Cornell Simple Equations, Deep Meaning #12;Introduction AI/Regression
Directed Regression Stanford University
Van Roy, Ben
Directed Regression Yi-hao Kao Stanford University Stanford, CA 94305 yihaokao Stanford, CA 94305 xyan@stanford.edu Abstract When used to guide decisions, linear regression analysis typically involves esti- mation of regression coefficients via ordinary least squares and their subsequent
The Augmented Complex Kernel LMS
Bouboulis, Pantelis; Mavroforakis, Michael
2011-01-01T23:59:59.000Z
Recently, a unified framework for adaptive kernel based signal processing of complex data was presented by the authors, which, besides offering techniques to map the input data to complex Reproducing Kernel Hilbert Spaces, developed a suitable Wirtinger-like Calculus for general Hilbert Spaces. In this short paper, the extended Wirtinger's calculus is adopted to derive complex kernel-based widely-linear estimation filters. Furthermore, we illuminate several important characteristics of the widely linear filters. We show that, although in many cases the gains from adopting widely linear estimation filters, as alternatives to ordinary linear ones, are rudimentary, for the case of kernel based widely linear filters significant performance improvements can be obtained.
Hypoelliptic heat kernel inequalities on H-type groups
Eldredge, Nathaniel Gilbert Bartsch
2009-01-01T23:59:59.000Z
Hypoelliptic heat kernel inequalities on the HeisenbergHypoelliptic heat kernel inequalities on Lie groups. PhDHypoelliptic heat kernel inequalities on H-type groups A
Slicing Regression: A Link-free Regression Method
Ker-Chau Li
2011-01-01T23:59:59.000Z
H. V. (1983). Reverse regression, fairness, and employmentleast squares linear regression. Statistica Sinica FISHER,A. S. (1981). Linear regression after selection. J.
Helices in High Dimensional Regression
Ker-Chau Li
2011-01-01T23:59:59.000Z
of "Sliced inverse regression." J. Amer. Statist. Assoc. 86C. (1992). Measurement error regression with unknown link:a projection-pursuit type regression model. Ann. Statist. 19
Principle of Least Squares Regression Equations Residuals Correlation and Regression
Watkins, Joseph C.
Principle of Least Squares Regression Equations Residuals Topic 3 Correlation and Regression Linear Regression I 1 / 15 #12;Principle of Least Squares Regression Equations Residuals Outline Principle of Least Squares Regression Equations Residuals 2 / 15 #12;Principle of Least Squares Regression Equations
Quantum regression theorem for non-Markovian Lindblad equations
Adrian A. Budini
2006-03-13T23:59:59.000Z
We find the conditions under which a quantum regression theorem can be assumed valid for non-Markovian master equations consisting in Lindblad superoperators with memory kernels. Our considerations are based on a generalized Born-Markov approximation, which allows us to obtain our results from an underlying Hamiltonian description. We demonstrate that a non-Markovian quantum regression theorem can only be granted in a stationary regime if the dynamics satisfies a quantum detailed balance condition. As an example we study the correlations of a two level system embedded in a complex structured reservoir and driven by an external coherent field.
Conventional regression models Unlabelled units
McCullagh, Peter
Conventional regression models Unlabelled units Consequences Sampling bias in logistic models Peter effects #12;Conventional regression models Unlabelled units Consequences Outline 1 Conventional regression models Gaussian models Binary regression model Properties of conventional models 2 Unlabelled units Point
Fast Kernel-Based Independent Component Analysis
McAuliffe, Jon
Fast Kernel-Based Independent Component Analysis Hao Shen , Stefanie Jegelka and Arthur Gretton instance, sources with near-zero kurtosis). FastKICA (Fast HSIC-based Kernel ICA) is a new optimisation-based ICA algorithms, FastKICA is applicable to any twice differentiable kernel function. Experimental
Regression from Local Features for Viewpoint and Pose Estimation Marwan Torki Ahmed Elgammal
Chen, Kuang-Yu
Regression from Local Features for Viewpoint and Pose Estimation Marwan Torki Ahmed Elgammal features in an image. The regression is learned from an embedded representation that reflects the local in computer vision can be formulated as regression problems where the goal is to learn a continu- ous real
B. Bruegmann
1993-12-02T23:59:59.000Z
The loop representation plays an important role in canonical quantum gravity because loop variables allow a natural treatment of the constraints. In these lectures we give an elementary introduction to (i) the relevant history of loops in knot theory and gauge theory, (ii) the loop representation of Maxwell theory, and (iii) the loop representation of canonical quantum gravity. (Based on lectures given at the 117. Heraeus Seminar, Bad Honnef, Sept. 1993)
QuestV: A Vi rtuali zed Multi kernel for Hi ghConfidence Systems Ye Li BostonU niversity ########## #### Matthew Danish BostonU niversity ######## #### Richard West BostonU niversity ############## #### Abstract operating together as a dis tributed system on a chip. QuestV uses virtualization techniques to isolate
Path Integral Representations on the Complex Sphere
Christian Grosche
2007-10-23T23:59:59.000Z
In this paper we discuss the path integral representations for the coordinate systems on the complex sphere S3C. The Schroedinger equation, respectively the path integral, separates in exactly 21 orthogonal coordinate systems. We enumerate these coordinate systems and we are able to present the path integral representations explicitly in the majority of the cases. In each solution the expansion into the wave-functions is stated. Also, the kernel and the corresponding Green function can be stated in closed form in terms of the invariant distance on the sphere, respectively on the hyperboloid.
Multivariate calibration with single-index signal regression Paul H.C. Eilers a
Marx, Brian D.
regression can be extended with an explicit link function between linear prediction and response is being estimated by P-splines. Application to simulations and three data sets shows that if a non-linearity from linear algebra, by non-linear functions. The idea is that a non-linear kernel in the linear space
Ordinary Least Square Regression, Orthogonal Regression, Geometric Mean Regression and their
Ordinary Least Square Regression, Orthogonal Regression, Geometric Mean Regression@notes.cc.sunysb.edu Abstract. Regression analysis, especially the ordinary least squares method which assumes that errors for both measurements. In this work, we examine two regression approaches available to accommodate
Robust Minimax Probability Machine Regression Robust Minimax Probability Machine Regression
Grudic, Greg
Robust Minimax Probability Machine Regression Robust Minimax Probability Machine Regression Thomas of Computer Science University of Colorado Boulder, C0 80309-0430, USA Abstract We formulate regression as maximizing the minimum probability () that the regression model is within ± of all future observations (i
Black Kernel and White Tip of Rice.
Martin, Alan L. (Alan La Mott); Altstatt, G. E. (George E.)
1940-01-01T23:59:59.000Z
on White Tip .......... 13 ................................. Recommendations for Control 14 Literature Cited .................................................. 14 BULLETIN NO. 584 MARCH 1940 BLACK KERNEL AND WHITE TIP OF RICE Alan L. Martin*, Plant...
Logistic regression Weakly informative priors
Gelman, Andrew
Logistic regression Weakly informative priors Conclusions Bayesian generalized linear models default p #12;Logistic regression Weakly informative priors Conclusions Classical logistic regression The problem of separation Bayesian solution Logistic regression -6 -4 -2 0 2 4 6 0.00.20.40.60.81.0 y = logit
Nonlinear regression Gordon K. Smyth
Smyth, Gordon K.
Nonlinear regression Gordon K. Smyth Volume 3, pp 14051411 in Encyclopedia of Environmetrics (ISBN, 2002 #12;Nonlinear regression The basic idea of nonlinear regression is the same as that of linear regression, namely to relate a response Y to a vector of predictor variables x D x1, . . . , xk T (see Linear
The Complex Gaussian Kernel LMS algorithm
Bouboulis, Pantelis
2010-01-01T23:59:59.000Z
Although the real reproducing kernels are used in an increasing number of machine learning problems, complex kernels have not, yet, been used, in spite of their potential interest in applications such as communications. In this work, we focus our attention on the complex gaussian kernel and its possible application in the complex Kernel LMS algorithm. In order to derive the gradients needed to develop the complex kernel LMS (CKLMS), we employ the powerful tool of Wirtinger's Calculus, which has recently attracted much attention in the signal processing community. Writinger's calculus simplifies computations and offers an elegant tool for treating complex signals. To this end, the notion of Writinger's calculus is extended to include complex RKHSs. Experiments verify that the CKLMS offers significant performance improvements over the traditional complex LMS or Widely Linear complex LMS (WL-LMS) algorithms, when dealing with nonlinearities.
transformations: representations
Nguyen, Dat H.
Overview 1. Number transformations: from one base to another 2. Integer representations 3. Real rate, caches... #12; ECS 50, Discussion on 4/25 2 Integer Transformation: From Decimal to Binary Let, Discussion on 4/25 3 Integer Transformation: From Binary to Decimal Compute the weight of each digit position
Comparing Methods for Multivariate Nonparametric Regression
Comparing Methods for Multivariate Nonparametric Regression David L. Banks \\Lambda Robert T, of the National Science Foundation or the U.S. government. #12; Keywords: multivariate nonparametric regression, linear regression, stepwise linear regression, additive models, AM, projection pursuit regression, PPR
LOG HAZARD REGRESSION Huiying Sun
Heckman, Nancy E.
LOG HAZARD REGRESSION by Huiying Sun Ph.D, Harbin Institute of Technology, Harbin, CHINA, 1991 regression splines to estimate the two log marginal hazard func tions of bivariate survival times, where, 1995) hazard regression for estimating a univariate survival time. We derive an approach to find
CORRESPONDENCE Reduced major axis regression
CORRESPONDENCE Reduced major axis regression and the island rule The Ôisland ruleŐ describes). The slope of the least-squares regression of Y on X, b, measures the extent to which large body sizes tend- ferent way, as the regression of Y ) X on X, and here the slope (b˘) is generally negative (b˘ = b ) 1
Linear regression issues in astronomy
Babu, G. Jogesh
Linear regression issues in astronomy Eric Feigelson Summer School in astrostatistics References regression Seeking the intrinsic relationship between two properties without specifying `dependent' and `independent' variables OLS(Y|X) OLS(X|Y) (inverse regr) Four symmetrical regression lines #12;Analytical
Simple Linear Regression Basic Ideas
Simple Linear Regression Basic Ideas Some Examples Least Squares Statistical View of Least Squares of Least Squares #12;Simple Linear Regression Basic Ideas Some Examples Least Squares Basic Ideas Suppose #12;Simple Linear Regression Basic Ideas Some Examples Least Squares Basic Ideas Suppose we have two
Kernelization and Enumeration: New Approaches to Solving Hard Problems
Meng, Jie
2011-08-08T23:59:59.000Z
their sizes. We present a 2k kernel for the cluster editing problem, which improves the previous best kernel of size 4k; We also present a linear kernel of size 7k 2d for the d-cluster editing problem, which is the first linear kernel for the problem...
ARCHITECTURE FOR A PORTABLE OPEN SOURCE REAL TIME KERNEL ENVIRONMENT
Lipari, Giuseppe
to the automotive industry. The set of tools comprises a schedulability analyzer, a kernel library and a kernel con platforms without sacri#28;cing performance. With our tools, the programmer only need to write of an Open Source Real Time Kernel for embedded automotive applications. Developing a kernel
Design of Positive-Definite Quaternion Kernels
Tobar, Felipe; Mandic, Danilo P.
2015-01-01T23:59:59.000Z
. Alpaydin, “Multiple kernel learning algorithms,” The Journal of Machine Learning Research, vol. 12, pp. 2211–2268, 2011. [7] F. Tobar, A. Kuh, and D. P. Mandic, “A novel augmented complex valued kernel LMS,” in Proc. of the 7th IEEE Sensor Array... and Multichannel Signal Processing Workshop, 2012, pp. 481–484. [8] P. Bouboulis, S. Theodoridis, and M. Mavroforakis, “The augmented complex kernel LMS,” IEEE Trans. on Signal Processing, vol. 60, no. 9, pp. 4962–4967, 2012. [9] F. Tobar and D. Mandic, “Quaternion...
Extension of Wirtinger's Calculus in Reproducing Kernel Hilbert Spaces and the Complex Kernel LMS
Bouboulis, Pantelis
2010-01-01T23:59:59.000Z
Over the last decade, kernel methods for nonlinear processing have successfully been used in the machine learning community. The primary mathematical tool employed in these methods is the notion of the Reproducing Kernel Hilbert Space. However, so far, the emphasis has been on batch techniques. It is only recently, that online techniques have been considered in the context of adaptive signal processing tasks. Moreover, these efforts have only been focussed on and real valued data sequences. To the best of our knowledge, no kernel-based strategy has been developed, so far, that is able to deal with complex valued signals. In this paper, we present a general framework to attack the problem of adaptive filtering of complex signals, using either real reproducing kernels, taking advantage of a technique called \\textit{complexification} of real RKHSs, or complex reproducing kernels, highlighting the use of the complex gaussian kernel. In order to derive gradients of operators that need to be defined on the associat...
Heat kernel asymptotics for magnetic Schrödinger operators
Bolte, Jens, E-mail: jens.bolte@rhul.ac.uk [Department of Mathematics, Royal Holloway, University of London, Egham TW20 0EX (United Kingdom)] [Department of Mathematics, Royal Holloway, University of London, Egham TW20 0EX (United Kingdom); Keppeler, Stefan, E-mail: stefan.keppeler@uni-tuebingen.de [Mathematisches Institut, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen (Germany)] [Mathematisches Institut, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen (Germany)
2013-11-15T23:59:59.000Z
We explicitly construct parametrices for magnetic Schrödinger operators on R{sup d} and prove that they provide a complete small-t expansion for the corresponding heat kernel, both on and off the diagonal.
Fractal Weyl law for Linux Kernel Architecture
L. Ermann; A. D. Chepelianskii; D. L. Shepelyansky
2010-09-16T23:59:59.000Z
We study the properties of spectrum and eigenstates of the Google matrix of a directed network formed by the procedure calls in the Linux Kernel. Our results obtained for various versions of the Linux Kernel show that the spectrum is characterized by the fractal Weyl law established recently for systems of quantum chaotic scattering and the Perron-Frobenius operators of dynamical maps. The fractal Weyl exponent is found to be $\
Complete regression of melanoma associated with vitiligo
Piqué-Duran, Enric; Palacios-Llopis, Santiago; Martínez-Martín, MŞSol; Pérez-Cejudo, Juan A
2011-01-01T23:59:59.000Z
melanoma: influence of regression. Histopathology 1983; 7:Varga E, Dobozy A. Tumor regression predicts higher risk ofcutaneous melanoma with regression doesnot require a lower
Regression with Missing X's: A Review
Roderick J. A. Little
2011-01-01T23:59:59.000Z
1968), "Missing Data in Regression Analysis," Journal of theTechnometrics, Little: Regression With Missing X's Ibrahim,Influence Multiple Linear Regression with Incomplete Data,"
Representation of Universal Algebra
Aleks Kleyn
2015-02-07T23:59:59.000Z
Theory of representations of universal algebra is a natural development of the theory of universal algebra. Morphism of the representation is the map that conserve the structure of the representation. Exploring of morphisms of the representation leads to the concepts of generating set and basis of representation. In the book I considered the notion of tower of representations of $F_i$-algebras, i=1 ..., n, as the set of coordinated representations of $F_i$-algebras.
SVMs with Profile-Based Kernels 1 Support Vector Machines with Profile-Based Kernels for
Pace, Gordon J.
amount of new protein sequences. The resulting sequences describe a protein in terms of the amino acids, the frequency that each amino acid appears in that column. Once a profile is available, a new sequence canSVMs with Profile-Based Kernels 1 Support Vector Machines with Profile-Based Kernels for Remote
Choquistic Regression: Generalizing Logistic Regression using the Choquet Integral
Hüllermeier, Eyke
to as choquistic regression, is to replace the linear function of pre- dictor variables, which is commonly used, it becomes possible to capture non-linear dependencies and in- teractions among predictor variables while regression, including the following ones: · Since the model is essentially linear in the in- put attributes
W.R. Tobler Bidimensional Regression
Tobler, Waldo
W.R. Tobler Bidimensional Regression Since its invention by Francis Galton in 1877 regression. Here the regression coefficients constitute a spatially varying, but coordinate invariant, second projection. In a computer implementation a nonparametric approach allows visualization of the regression
Facial Action Unit Intensity Prediction via Hard Multi-Task Metric Learning for Kernel Regression
]), characterizing activations of face muscles (AUs) for describing facial expressions. In order to be able, all related to mouth movements. Thus, some features are potentially relevant for all five tasks
Understanding Kernel Ridge Regression: Common behaviors from simple functions to density functionals
Vu, Kevin; Li, Li; Rupp, Matthias; Chen, Brandon F; Khelif, Tarek; Müller, Klaus-Robert; Burke, Kieron
2015-01-01T23:59:59.000Z
Accurate approximations to density functionals have recently been obtained via machine learning (ML). By applying ML to a simple function of one variable without any random sampling, we extract the qualitative dependence of errors on hyperparameters. We find universal features of the behavior in extreme limits, including both very small and very large length scales, and the noise-free limit. We show how such features arise in ML models of density functionals.
Olive, David
1D Regression David J. Olive Southern Illinois University August 27, 2004 Abstract Regression regression, Y is independent of x given a single linear combination +T x of the predictors. Special cases of 1D regression include multiple linear regression, logistic regression, generalized linear models
Fast generation of sparse random kernel graphs
DOE Public Access Gateway for Energy & Science Beta (PAGES Beta)
Hagberg, Aric; Lemons, Nathan; Du, Wen -Bo
2015-09-10T23:59:59.000Z
The development of kernel-based inhomogeneous random graphs has provided models that are flexible enough to capture many observed characteristics of real networks, and that are also mathematically tractable. We specify a class of inhomogeneous random graph models, called random kernel graphs, that produces sparse graphs with tunable graph properties, and we develop an efficient generation algorithm to sample random instances from this model. As real-world networks are usually large, it is essential that the run-time of generation algorithms scales better than quadratically in the number of vertices n. We show that for many practical kernels our algorithm runs in timemore »at most ?(n(logn)˛). As an example, we show how to generate samples of power-law degree distribution graphs with tunable assortativity.« less
A dynamic kernel modifier for linux
Minnich, R. G. (Ronald G.)
2002-09-03T23:59:59.000Z
Dynamic Kernel Modifier, or DKM, is a kernel module for Linux that allows user-mode programs to modify the execution of functions in the kernel without recompiling or modifying the kernel source in any way. Functions may be traced, either function entry only or function entry and exit; nullified; or replaced with some other function. For the tracing case, function execution results in the activation of a watchpoint. When the watchpoint is activated, the address of the function is logged in a FIFO buffer that is readable by external applications. The watchpoints are time-stamped with the resolution of the processor high resolution timers, which on most modem processors are accurate to a single processor tick. DKM is very similar to earlier systems such as the SunOS trace device or Linux TT. Unlike these two systems, and other similar systems, DKM requires no kernel modifications. DKM allows users to do initial probing of the kernel to look for performance problems, or even to resolve potential problems by turning functions off or replacing them. DKM watchpoints are not without cost: it takes about 200 nanoseconds to make a log entry on an 800 Mhz Pentium-Ill. The overhead numbers are actually competitive with other hardware-based trace systems, although it has less 'Los Alamos National Laboratory is operated by the University of California for the National Nuclear Security Administration of the United States Department of Energy under contract W-7405-ENG-36. accuracy than an In-Circuit Emulator such as the American Arium. Once the user has zeroed in on a problem, other mechanisms with a higher degree of accuracy can be used.
Experimental study of turbulent flame kernel propagation
Mansour, Mohy [National Institute of Laser Enhanced Sciences, Cairo University, Giza (Egypt); Peters, Norbert; Schrader, Lars-Uve [Institute of Combustion Technology, Aachen (Germany)
2008-07-15T23:59:59.000Z
Flame kernels in spark ignited combustion systems dominate the flame propagation and combustion stability and performance. They are likely controlled by the spark energy, flow field and mixing field. The aim of the present work is to experimentally investigate the structure and propagation of the flame kernel in turbulent premixed methane flow using advanced laser-based techniques. The spark is generated using pulsed Nd:YAG laser with 20 mJ pulse energy in order to avoid the effect of the electrodes on the flame kernel structure and the variation of spark energy from shot-to-shot. Four flames have been investigated at equivalence ratios, {phi}{sub j}, of 0.8 and 1.0 and jet velocities, U{sub j}, of 6 and 12 m/s. A combined two-dimensional Rayleigh and LIPF-OH technique has been applied. The flame kernel structure has been collected at several time intervals from the laser ignition between 10 {mu}s and 2 ms. The data show that the flame kernel structure starts with spherical shape and changes gradually to peanut-like, then to mushroom-like and finally disturbed by the turbulence. The mushroom-like structure lasts longer in the stoichiometric and slower jet velocity. The growth rate of the average flame kernel radius is divided into two linear relations; the first one during the first 100 {mu}s is almost three times faster than that at the later stage between 100 and 2000 {mu}s. The flame propagation is slightly faster in leaner flames. The trends of the flame propagation, flame radius, flame cross-sectional area and mean flame temperature are related to the jet velocity and equivalence ratio. The relations obtained in the present work allow the prediction of any of these parameters at different conditions. (author)
Crystal Structure Representations for Machine Learning Models of Formation Energies
Faber, Felix; von Lilienfeld, O Anatole; Armiento, Rickard
2015-01-01T23:59:59.000Z
We introduce and evaluate a set of feature vector representations of crystal structures for machine learning (ML) models of formation energies of solids. ML models of atomization energies of organic molecules have been successful using a Coulomb matrix representation of the molecule. We consider three ways to generalize such representations to periodic systems: (i) a matrix where each element is related to the Ewald sum of the electrostatic interaction between two different atoms in the unit cell repeated over the lattice; (ii) an extended Coulomb-like matrix that takes into account a number of neighboring unit cells; and (iii) an Ansatz that mimics the periodicity and the basic features of the elements in the Ewald sum matrix by using a sine function of the crystal coordinates of the atoms. The representations are compared for a Laplacian kernel with Manhattan norm, trained to reproduce formation energies using a data set of 3938 crystal structures obtained from the Materials Project. For training sets consi...
On fusion kernel in Liouville theory
Nikita Nemkov
2014-09-29T23:59:59.000Z
We study fusion kernel for non-degenerate conformal blocks in Liouville theory as a solution to the difference equations originating from the pentagon identity. We suggest an approach to these equations based on 'non-perturbative' series expansion which allows to calculate the fusion kernel iteratively. We also find the exact solutions for the cases when the central charge is $c=1+6(b-b^{-1})^2$ and $b~\\in \\mathbb{N}$. For $c = 1$ our result reproduces the formula, obtained earlier from analytical continuation via Painlev\\'e equation. However, in our case it appears in a significantly simplified form.
Is the gasoline tax regressive?
Poterba, James M.
1990-01-01T23:59:59.000Z
Claims of the regressivity of gasoline taxes typically rely on annual surveys of consumer income and expenditures which show that gasoline expenditures are a larger fraction of income for very low income households than ...
Measurements of the Thermal Neutron Scattering Kernel
Danon, Yaron
Measurements of the Thermal Neutron Scattering Kernel Li (Emily) Liu, Yaron Danon, Bjorn Becker and discussions Problems and Future study Questions #12;3 M. Mattes and J. Keinert, Thermal Neutron Scattering experimental data used was from 1973-1974! M. Mattes and J. Keinert, Thermal Neutron Scattering Data
Nonextensive Entropic Kernels Andre F. T. Martins
Aguiar, Pedro M. Q.
. Some of these kernels are related to classic infor- mation theoretic quantities, such as mutual information and the Jensen-Shannon diver- gence. Meanwhile, driven by recent advances in Tsallis statistics on Machine Learning, Helsinki, Finland, 2008. Copy- right 2008 by the author(s)/owner(s). approaches that map
Choosing a Kernel for Cross-Validation
Savchuk, Olga
2010-01-14T23:59:59.000Z
methods of bandwidth selection termed: Indirect cross-validation and Robust one-sided cross- validation. The kernels used in the Indirect cross-validation method yield an improvement in the relative bandwidth rate to n^1=4, which is substantially better...
Comparing Methods for Multivariate Nonparametric Regression
Comparing Methods for Multivariate Nonparametric Regression David L. Banks Robert T. Olszewskiy Roy, of the National Science Foundation or the U.S. government. #12;Keywords: multivariate nonparametric regression, linear regression, stepwise linear regression, additive models, AM, projection pursuit regression, PPR
Conventional regression models Auto-generated units
McCullagh, Peter
Conventional regression models Auto-generated units Consequences of auto-generation Inference Royal Statistical Society Feb 6, 2008 Peter McCullagh Auto-generated units #12;Conventional regression Conventional regression models Gaussian models Binary regression model Properties of regression models Problems
Conventional regression models Auto-generated units
McCullagh, Peter
Conventional regression models Auto-generated units Consequences of auto-generation Inference regression models Auto-generated units Consequences of auto-generation Inference and prediction Outline 1 Conventional regression models Gaussian models Binary regression model Properties of regression models Problems
A general approach to statistical modeling of physical laws: nonparametric regression
I. Grabec
2007-04-01T23:59:59.000Z
Statistical modeling of experimental physical laws is based on the probability density function of measured variables. It is expressed by experimental data via a kernel estimator. The kernel is determined objectively by the scattering of data during calibration of experimental setup. A physical law, which relates measured variables, is optimally extracted from experimental data by the conditional average estimator. It is derived directly from the kernel estimator and corresponds to a general nonparametric regression. The proposed method is demonstrated by the modeling of a return map of noisy chaotic data. In this example, the nonparametric regression is used to predict a future value of chaotic time series from the present one. The mean predictor error is used in the definition of predictor quality, while the redundancy is expressed by the mean square distance between data points. Both statistics are used in a new definition of predictor cost function. From the minimum of the predictor cost function, a proper number of data in the model is estimated.
Testing regression models with fewer regressors
Judea Pearl; P. Meshkat
2011-01-01T23:59:59.000Z
Testing Regression Models W i t h Fewer Regressors JudeaINTRODUCTION Let M be a recursive regression model, that is,a set of regression equations for ordered variables Yi, ,
Review: Logistic regression, Gaussian nave Bayes, linear regression, and their connections
Mitchell, Tom
Review: Logistic regression, Gaussian naďve Bayes, linear regression, and their connections Yi, and feature selection #12;Outline Logistic regression Decision surface (boundary) of classifiers Generative vs. discriminative classifiers Linear regression Bias-variance decomposition and tradeoff
Marcus Hutter -1 -Bayesian Regression of Piecewise Constant Functions Bayesian Regression of
Hutter, Marcus
Marcus Hutter - 1 - Bayesian Regression of Piecewise Constant Functions Bayesian Regression6 June 2006 #12;Marcus Hutter - 2 - Bayesian Regression of Piecewise Constant Functions Table of Contents · Bayesian Regression · Quantities of Interest · Efficient Solutions by Dynamic Programming · Determination
Broader source: All U.S. Department of Energy (DOE) Office Webpages (Extended Search)
representation of this product. Ron Brightwell Joint entry Operating Systems Research 1527 16th NW 5 Washington, DC 20036 USA Trammell Hudson Phone:...
Regression analysis with missing data
Michelli, Frank Anthony
1968-01-01T23:59:59.000Z
: Statistios REGRESSION ANALYSIS WITH MISS1NG DATA A Thesis FRANK ANTHONY MICHELLI Approved as to style and content by: hairman of o ttee Member Head of Department Member Member Zanuary 196B ACZNOWLED ONE NT S I can only begin to express my sincere...
Simultaneous regression and clustering to predict movie ratings
Rodriguez, Matthew
2010-01-01T23:59:59.000Z
Testing the Regression Coefficients . . . . . . . . .2 Performing Logistic Regression on the Movielens dataset2.4 Logistic Regression Experiments . . . . . . . . . .
SMOOTHED ESTIMATING EQUATIONS FOR INSTRUMENTAL VARIABLES QUANTILE REGRESSION
Kaplan, David M.; Sun, Yixiao
2012-01-01T23:59:59.000Z
Instrumental quantile regression inference for structuralsample inference for quantile regression models. Journal ofmethods for median regression models. Econometrica,
Visualizing 1D Regression David J. Olive
Olive, David
regression, binary regression and general- ized linear models. If a good estimate ^b of some non a single linear combination T x of the predictors. Special cases of 1D regression include multiple linear(y) = + T x + e. Generalized linear models (GLM's) are also a special case of 1D regression. Some notation
On Robust Regression in Photogrammetric Point Clouds
Schindler, Konrad
On Robust Regression in Photogrammetric Point Clouds Konrad Schindler and Horst Bischof Institute,bischof}@icg.tu-graz.ac.at Abstract. Many applications in computer vision require robust linear regression on photogrammetrically for robust regression are based on distance measures from the regression surface to the points
Regression models Consequences of auto-generation
McCullagh, Peter
Regression models Consequences of auto-generation All creatures great and small Inference University of Chicago Statslab Cambridge Nov 14 2008 Peter McCullagh Auto-generated units #12;Regression Regression models Gaussian models Binary regression model Attenuation of treatment effect Problems
Conventional regression models Auto-generated units
McCullagh, Peter
Conventional regression models Auto-generated units Consequences of auto-generation Inference regression models Auto-generated units Consequences of auto-generation Inference and prediction Outline 1 Conventional regression models Gaussian models Binary regression model Attenuation of treatment effect Problems
Conventional regression models Auto-generated units
McCullagh, Peter
Conventional regression models Auto-generated units Consequences of auto-generation Arguments pro-generated units #12;Conventional regression models Auto-generated units Consequences of auto-generation Arguments pro and con Outline 1 Conventional regression models Gaussian models Binary regression model
36-220 Lab #11 Multiple Regression
Spirtes, Peter
36-220 Lab #11 Multiple Regression Please write your name below, tear off this front page and give;36-220 Lab #11 Multiple Regression First let's download the dataset we will need for the lab. Download gas Simple Regression Analysis 1. Fit the least squares regression line to describe the relationship between
A Formulation for Minimax Probability Machine Regression
Grudic, Greg
A Formulation for Minimax Probability Machine Regression Thomas Strohmann Department of Computer Science University of Colorado, Boulder grudic@cs.colorado.edu Abstract We formulate the regression of the regression model will be within some ± bound of the true regression function. Our formulation is unique
ContentsContents4343Regression and correlation
Vickers, James
ContentsContents4343Regression and correlation 1. Regression 2. Correlation Learning outcomes You. 1 #12;Regression 43.1 Introduction Problems in engineering often involve the exploration of the relationship(s) between two or more variables. The technique of regression analysis is very useful and well
On admissible memory kernels for random unitary qubit evolution
Filip A. Wudarski; Pawe? Nale?yty; Gniewomir Sarbicki; Dariusz Chru?ci?ski
2015-04-12T23:59:59.000Z
We analyze random unitary evolution of the qubit within memory kernel approach. We provide su?cient conditions which guarantee that the corresponding memory kernel generates physically legitimate quantum evolution. Interestingly, we are able to recover several well known examples and generate new classes of nontrivial qubit evolution. Surprisingly, it turns out that quantum evolution with memory kernel generated by our approach gives rise to vanishing non-Markovianity measure based on the distinguishability of quantum states.
U-086:Linux Kernel "/proc//mem" Privilege Escalation Vulnerability
Broader source: Energy.gov [DOE]
A vulnerability has been discovered in the Linux Kernel, which can be exploited by malicious, local users to gain escalated privileges.
U-175: Linux Kernel KVM Memory Slot Management Flaw
Broader source: Energy.gov [DOE]
A vulnerability was reported in the Linux Kernel. A local user on the guest operating system can cause denial of service conditions on the host operating system.
Regression analysis with longitudinal measurements
Ryu, Duchwan
2005-08-29T23:59:59.000Z
, in the cardiotoxic effects of doxorubicin chemotherapy for the treat- ment of acute lymphoblastic leukemia in childhood (Lipsitz et al., 2002; Fitzmaurice et al., 2003), the design points are not pre-defined but determined by the preceding response. This outcome...-dependent feature of measurements makes biased estimation of regression line. As noticed by Lipsitz et al. (2002); Fitzmaurice et al. (2003), even the least square estimates will be biased, which does not require the distributional assumption of response error...
Partial least squares regression as an alternative to current regression methods used in ecology
Carrascal, Luis M.
Partial least squares regression as an alternative to current regression methods used in ecology regression analysis (PLSR), and its potential utility in ecological studies. This statistical technique with multiple regression (MR) and with a combination of principal component analysis and multiple regression
Shadow Kernels: A General Mechanism For Kernel Specialization in Existing Operating Systems
Chick, Oliver R. A.; Carata, Lucian; Snee, James; Balakrishnan, Nikilesh; Sohan, Ripduman
2015-01-01T23:59:59.000Z
in systems such as Ar- rakis [13, 14]. Once a process is given ownership of a particular vir- tualized PCI device, the problem of “routing” asynch- ronous tasks towards executing a given shadow kernel becomes solvable: All the code that is executed in ker...
Shape Recipes: Scene Representations that Refer to the Image
Torralba, Antonio
Shape Recipes: Scene Representations that Refer to the Image William T. Freeman and Antonio- resentation, called a scene recipe, that relies on the image itself to de- scribe the complex scene configurations. Shape recipes are an example: these are the regression coefficients that predict the bandpassed
Regression Analysis with an Unknown Link Function: the Adjoint Projection Pursuit Regression
Naihua Duan
2011-01-01T23:59:59.000Z
Misspecified Non- linear Regression Models," Journal of theAdjoint Projection Pursuit Regression also give examples inL is the estimated linear regression of x on bx. Note Remark
Extended Kalman Filter Using a Kernel Recursive Least Squares Observer
Slatton, Clint
Extended Kalman Filter Using a Kernel Recursive Least Squares Observer Pingping Zhu, Badong Chen estimation problem combining the extended Kalman filter (EKF) with a kernel recursive least squares (KRLS Kalman filter, EKF and KRLS algorithms. Results demonstrate that the performance of the EKF
Structured Linux Kernel Projects for Teaching Operating Systems Concepts
Laadan, Oren
Structured Linux Kernel Projects for Teaching Operating Systems Concepts Oren Laadan Dept of a complex, production operating system can be a challenge. We present a structured series of five Linux kernel programming projects suitable for a one semester introductory operating systems course to address
CDF and Survival Function Estimation with Infinite-Order Kernels
Politis, Dimitris N.
) and the survival function is proposed using infinite-order kernels. Fourier transform theory on generalizedCDF and Survival Function Estimation with Infinite-Order Kernels Arthur Berg and Dimitris N sample sizes these estimators can significantly improve the estimation of the CDF and survival function
FRAME BASED KERNEL METHODS FOR AUTOMATIC CLASSIFICATION IN HYPERSPECTRAL DATA
Hirn, Matthew
FRAME BASED KERNEL METHODS FOR AUTOMATIC CLASSIFICATION IN HYPERSPECTRAL DATA John J. Benedetto, instead of (orthonormal) bases. Our frames are data-dependent and are based on endmember demixing schemes propose a new kernel and frame based dimension reduc- ing algorithm by exploiting the synergy between
Mach Kernel Monitor (with applications using the PIE environment)
Mach Kernel Monitor (with applications using the PIE environment) 23 August 1990 Ted Lehr David. Special examples of data obtained by using MKM are shown via the PIE performance monitoring environment programs are scheduled. This manual describes how to use MKM, the Mach (contextswitch) kernel monitor
Scientific Computing Kernels on the Cell Processor
Williams, Samuel W.; Shalf, John; Oliker, Leonid; Kamil, Shoaib; Husbands, Parry; Yelick, Katherine
2007-04-04T23:59:59.000Z
The slowing pace of commodity microprocessor performance improvements combined with ever-increasing chip power demands has become of utmost concern to computational scientists. As a result, the high performance computing community is examining alternative architectures that address the limitations of modern cache-based designs. In this work, we examine the potential of using the recently-released STI Cell processor as a building block for future high-end computing systems. Our work contains several novel contributions. First, we introduce a performance model for Cell and apply it to several key scientific computing kernels: dense matrix multiply, sparse matrix vector multiply, stencil computations, and 1D/2D FFTs. The difficulty of programming Cell, which requires assembly level intrinsics for the best performance, makes this model useful as an initial step in algorithm design and evaluation. Next, we validate the accuracy of our model by comparing results against published hardware results, as well as our own implementations on a 3.2GHz Cell blade. Additionally, we compare Cell performance to benchmarks run on leading superscalar (AMD Opteron), VLIW (Intel Itanium2), and vector (Cray X1E) architectures. Our work also explores several different mappings of the kernels and demonstrates a simple and effective programming model for Cell's unique architecture. Finally, we propose modest microarchitectural modifications that could significantly increase the efficiency of double-precision calculations. Overall results demonstrate the tremendous potential of the Cell architecture for scientific computations in terms of both raw performance and power efficiency.
Regression quantiles for time series
Cai, Zongwu
2002-02-01T23:59:59.000Z
~see, e+g+, Ibragimov and Linnik, 1971, p+ 316!+ Namely, partition REGRESSION QUANTILES FOR TIME SERIES 187 $1, + + + , n% into 2qn 1 1 subsets with large block of size r 5 rn and small block of size s 5 sn+ Set q 5 qn 5 ? n rn 1 sn? , (A.7) where {x...! are the standard Lindeberg–Feller conditions for asymptotic normality of Qn,1 for the independent setup+ Let us first establish ~A+8!+ To this effect, we define the large-block size rn by rn 5 {~nhn!102} and the small-block size sn 5 {~nhn!1020log n}+ Then, as n r...
Regression analysis of cytopathological data
Whittemore, A.S.; McLarty, J.W.; Fortson, N.; Anderson, K.
1982-12-01T23:59:59.000Z
Epithelial cells from the human body are frequently labelled according to one of several ordered levels of abnormality, ranging from normal to malignant. The label of the most abnormal cell in a specimen determines the score for the specimen. This paper presents a model for the regression of specimen scores against continuous and discrete variables, as in host exposure to carcinogens. Application to data and tests for adequacy of model fit are illustrated using sputum specimens obtained from a cohort of former asbestos workers.
Thresholding Multivariate Regression and Generalized Principal Components
Sun, Ranye
2014-03-17T23:59:59.000Z
the curse of dimensionality. It is desirable to estimate the regression coefficient matrix by low-rank matrices constructed from its SVD. We reduce such regression problems to sparse SVD problems for cor- related data matrices and generalize the fast...
Batch-mode Supervised Learning Linear regression
Wehenkel, Louis
Batch-mode Supervised Learning Linear regression Applied inductive learning - Lecture 3 Louis (& Pierre Geurts)AIA... (1/19) #12;Batch-mode Supervised Learning Linear regression Batch-mode Supervised Learning Linear regression Least mean square error solution Regularization and algorithmics Residual
Exact and Approximate REML for Heteroscedastic Regression
Smyth, Gordon K.
Exact and Approximate REML for Heteroscedastic Regression Gordon K. Smyth Department of Mathematics, the above het- eroscedastic regression model is the most general model of the type considered by LN98 and SV to estimate the het- eroscedastic regression model by way of two coupled generalized linear models
Nonparametric Regression with Correlated Errors Jean Opsomer
Wang, Yuedong
Nonparametric Regression with Correlated Errors Jean Opsomer Iowa State University Yuedong Wang Nonparametric regression techniques are often sensitive to the presence of correlation in the errors splines and wavelet regression under correlation, both for short-range and long-range dependence
Regression on feature projections H. Altay Guvenir
Güvenir, H. Altay
Regression on feature projections H. Altay Guvenir * Uysal a Department Computer Engineering ######## This paper describes machine learning method, called Regression Feature Projections (RFP), predicting real with KNN based-regression algorithms. Results real sets achieves or comparable accuracy and is both Rule
ENGI 3423 Simple Linear Regression Page 12-01 Simple Linear Regression
George, Glyn
for dealing with non-linear regression are available in the course text, but are beyond the scopeENGI 3423 Simple Linear Regression Page 12-01 Simple Linear Regression Sometimes an experiment predict the value of Y for that value of x . The simple linear regression model is that the predicted
1D Regression ... estimates of the linear regression coefficients are relevant to the linear
Olive, David
Chapter 12 1D Regression ... estimates of the linear regression coefficients are relevant and look for a functional form for g(·). Brillinger (1983, p. 98) Regression is the study. The scalar Y is a random variable and x is a random vector. A special case of regression is multiple linear
1D Regression ... estimates of the linear regression coefficients are relevant to the linear
Olive, David
Chapter 15 1D Regression ... estimates of the linear regression coefficients are relevant and look for a functional form for g(·). Brillinger (1983, p. 98) Regression is the study. The scalar Y is a random variable and x is a random vector. A special case of regression is multiple linear
Robust Regression Appendix to An R and S-PLUS Companion to Applied Regression
Masci, Frank
Robust Regression Appendix to An R and S-PLUS Companion to Applied Regression John Fox January 2002 regression, is to employ a fitting criterion that is not as vulnerable as least squares to unusual data. The most common general method of robust regression is M-estimation, introduced by Huber (1964).1 Consider
Scaling regression inputs by dividing by 2 sd Prior distribution for logistic regression
Gelman, Andrew
Scaling regression inputs by dividing by 2 sd Prior distribution for logistic regression Comparing Andrew Gelman Some Recent Progress in Simple Statistical Methods #12;Scaling regression inputs by dividing by 2 sd Prior distribution for logistic regression Comparing the upper third to the lower third
The Dynamical Kernel Scheduler - Part 1
Adelmann, Andreas; Suter, Andreas
2015-01-01T23:59:59.000Z
Emerging processor architectures such as GPUs and Intel MICs provide a huge performance potential for high performance computing. However developing software using these hardware accelerators introduces additional challenges for the developer such as exposing additional parallelism, dealing with different hardware designs and using multiple development frameworks in order to use devices from different vendors. The Dynamic Kernel Scheduler (DKS) is being developed in order to provide a software layer between host application and different hardware accelerators. DKS handles the communication between the host and device, schedules task execution, and provides a library of built-in algorithms. Algorithms available in the DKS library will be written in CUDA, OpenCL and OpenMP. Depending on the available hardware, the DKS can select the appropriate implementation of the algorithm. The first DKS version was created using CUDA for the Nvidia GPUs and OpenMP for Intel MIC. DKS was further integrated in OPAL (Object-or...
Regression analysis for peak designation in pulsatile pressure signals
Scalzo, Fabien; Xu, Peng; Asgari, Shadnaz; Bergsneider, Marvin; Hu, Xiao
2009-01-01T23:59:59.000Z
5 ORIGINAL ARTICLE Regression analysis for peak designationwith more versatile regression models. The experimentalof different state-of-the-art regression analysis methods is
RegTools: A Julia Package for Assisting Regression Analysis
Liang, Muzhou
2015-01-01T23:59:59.000Z
and R. E. Welsch. 1980. ”Regression Diagnostics: IdentifyingObservation in Linear Regression. ” Techometrics 19:15-18. [Observations in Linear Regression. ” Journal of the American
Automating approximate Bayesian computation by local linear regression
Thornton, Kevin R
2009-01-01T23:59:59.000Z
computation by local linear regression Kevin R Thorntonof ABC based on using a linear regression to approximate theimplements the local linear-regression approach to ABC. The
Kernels for Feedback Arc Set In Tournaments Stephane Bessy
Paris-Sud XI, Université de
Kernels for Feedback Arc Set In Tournaments St´ephane Bessy Fedor V. Fomin Serge Gaspers Christophe´e de Montpellier 2, CNRS, 161 rue Ada, 34392 Montpellier, France. {bessy
Efficient graphlet kernels for large graph comparison Nino Shervashidze
Mehlhorn, Kurt
Efficient graphlet kernels for large graph comparison Nino Shervashidze MPI for Biological such as gSpan (Yan & Han, 2003) have been developed for this task, which use el- egant data structures
KALMAN FILTERING IN REPRODUCING KERNEL HILBERT SPACES PINGPING ZHU
Slatton, Clint
KALMAN FILTERING IN REPRODUCING KERNEL HILBERT SPACES By PINGPING ZHU A DISSERTATION PRESENTED.1.1 Bayesian Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.1.2 Kalman Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.1.3 Nonlinear Kalman Filter . . . . . . . . . . . . . . . . . . . . . . . . 19 2.1.3.1 Extended
MA699: Stochastic Taylor expansions and heat kernel asymptotics
2011-01-17T23:59:59.000Z
Jan 17, 2011 ... where p : (0,+?) × Rn × Rn ? R is a smooth function that is called the heat kernel ... Theorem 2.1 Let us assume that b and ? are smooth, and that their ..... [
complexity of the classical kernel functions of potential theory
1998-08-27T23:59:59.000Z
1991 Mathematics Subject Classification. ...... It is a simple exercise using the argument principle that the ... off the following facts (keep in mind that wk is close to An ? ?n). ...... excellent strategy for computing the Poisson kernel efficiently.
RECIPES FOR CLASSICAL KERNEL FUNCTIONS ASSOCIATED TO A
RECIPES FOR CLASSICAL KERNEL FUNCTIONS ASSOCIATED TO A MULTIPLY CONNECTED DOMAIN IN THE PLANE elementary functions that are easy to compute. I shall also give a recipe to compute the functions
Green's kernels for transmission problems in bodies with small inclusions
Vladimir Maz'ya; Alexander Movchan; Michael Nieves
2010-05-24T23:59:59.000Z
The uniform asymptotic approximation of Green's kernel for the transmission problem of antiplane shear is obtained for domains with small inclusions. The remainder estimates are provided. Numerical simulations are presented to illustrate the effectiveness of the approach.
Polymer representations and geometric quantization
Miguel Campiglia
2011-11-02T23:59:59.000Z
Polymer representations of the Weyl algebra of linear systems provide the simplest analogues of the representation used in loop quantum gravity. The construction of these representations is algebraic, based on the Gelfand-Naimark-Segal construction. Is it possible to understand these representations from a Geometric Quantization point of view? We address this question for the case of a two dimensional phase space.
Modeling Multiple Drugs on Lung Cancer and Normal Cells using Regression
Chen, Michelle
2013-01-01T23:59:59.000Z
GLM: Binomial Regression Model . . . . . . . . . .Linear Regression: Least SquaresCourse in Statistics Regression Analysis. Pearson Education,
Varying-Coefficient Functional Linear Regression Models
Cardot, Hervé
Varying-Coefficient Functional Linear Regression Models Herv´e Cardot1 and Pascal Sarda2 1, the ability of such non linear functional approaches to produce competitive estimations. Short title : Varying monograph. We propose here another generalization of the functional linear regression model in which
Computational Reality XIII Non-linear regression
Berlin,Technische Universität
Computational Reality XIII Non-linear regression Inverse analysis II B. Emek Abali @ LKM - TU Berlin Abstract Linear regression to fit and determine parameters, shown in the last tutorial, is quite useful and widely implemented, however, there are material models where parameters are coupled non-linearly
Regression of Environmental Noise in LIGO Data
Vaibhav Tiwari; Marco Drago; Valery Frolov; Sergey Klimenko; Guenakh Mitselmakher; Valentin Necula; Giovanni Prodi; Virginia Re; Francesco Salemi; Gabriele Vedovato; Igor Yakushin
2015-03-25T23:59:59.000Z
We address the problem of noise regression in the output of gravitational-wave (GW) interferometers, using data from the physical environmental monitors (PEM). The objective of the regression analysis is to predict environmental noise in the gravitational-wave channel from the PEM measurements. One of the most promising regression method is based on the construction of Wiener-Kolmogorov filters. Using this method, the seismic noise cancellation from the LIGO GW channel has already been performed. In the presented approach the Wiener-Kolmogorov method has been extended, incorporating banks of Wiener filters in the time-frequency domain, multi-channel analysis and regulation schemes, which greatly enhance the versatility of the regression analysis. Also we presents the first results on regression of the bi-coherent noise in the LIGO data.
Regression of Environmental Noise in LIGO Data
Tiwari, Vaibhav; Frolov, Valery; Klimenko, Sergey; Mitselmakher, Guenakh; Necula, Valentin; Prodi, Giovanni; Re, Virginia; Salemi, Francesco; Vedovato, Gabriele; Yakushin, Igor
2015-01-01T23:59:59.000Z
We address the problem of noise regression in the output of gravitational-wave (GW) interferometers, using data from the physical environmental monitors (PEM). The objective of the regression analysis is to predict environmental noise in the gravitational-wave channel from the PEM measurements. One of the most promising regression method is based on the construction of Wiener-Kolmogorov filters. Using this method, the seismic noise cancellation from the LIGO GW channel has already been performed. In the presented approach the Wiener-Kolmogorov method has been extended, incorporating banks of Wiener filters in the time-frequency domain, multi-channel analysis and regulation schemes, which greatly enhance the versatility of the regression analysis. Also we presents the first results on regression of the bi-coherent noise in the LIGO data.
Huang, Su-Yun
Multiclass Support Vector Classification via Regression Multiclass Support Vector Classification via Regression Pei-Chun Chen peichun@stat.sinica.edu.tw Institute of Statistical Science Academia classification is considered and resolved through the mul- tiresponse linear regression approach. Scores are used
Stout, Quentin F.
2008-01-01T23:59:59.000Z
In Computational Statistics and Data Analysis 53 (2008), pp. 289297 Unimodal Regression via Prefix Isotonic Regression Quentin F. Stout University of Michigan Ann Arbor, MI 481092121 Abstract This paper gives algorithms for determining real-valued uni- variate unimodal regressions, that is, for determining
On flame kernel formation and propagation in premixed gases
Eisazadeh-Far, Kian; Metghalchi, Hameed [Northeastern University, Mechanical and Industrial Engineering Department, Boston, MA 02115 (United States); Parsinejad, Farzan [Chevron Oronite Company LLC, Richmond, CA 94801 (United States); Keck, James C. [Massachusetts Institute of Technology, Cambridge, MA 02139 (United States)
2010-12-15T23:59:59.000Z
Flame kernel formation and propagation in premixed gases have been studied experimentally and theoretically. The experiments have been carried out at constant pressure and temperature in a constant volume vessel located in a high speed shadowgraph system. The formation and propagation of the hot plasma kernel has been simulated for inert gas mixtures using a thermodynamic model. The effects of various parameters including the discharge energy, radiation losses, initial temperature and initial volume of the plasma have been studied in detail. The experiments have been extended to flame kernel formation and propagation of methane/air mixtures. The effect of energy terms including spark energy, chemical energy and energy losses on flame kernel formation and propagation have been investigated. The inputs for this model are the initial conditions of the mixture and experimental data for flame radii. It is concluded that these are the most important parameters effecting plasma kernel growth. The results of laminar burning speeds have been compared with previously published results and are in good agreement. (author)
Kernel density estimation of a multidimensional efficiency profile
Anton Poluektov
2014-11-20T23:59:59.000Z
Kernel density estimation is a convenient way to estimate the probability density of a distribution given the sample of data points. However, it has certain drawbacks: proper description of the density using narrow kernels needs large data samples, whereas if the kernel width is large, boundaries and narrow structures tend to be smeared. Here, an approach to correct for such effects, is proposed that uses an approximate density to describe narrow structures and boundaries. The approach is shown to be well suited for the description of the efficiency shape over a multidimensional phase space in a typical particle physics analysis. An example is given for the five-dimensional phase space of the $\\Lambda_b^0\\to D^0p\\pi$ decay.
A Library for Locally Weighted Projection Regression
Klanke, Stefan; Vijayakumar, Sethu; Schaal, Stefan
2008-01-01T23:59:59.000Z
In this paper we introduce an improved implementation of locally weighted projection regression (LWPR), a supervised learning algorithm that is capable of handling high-dimensional input data. As the key features, our ...
Spatial variation decomposition via sparse regression
Zhang, Wangyang
In this paper, we briefly discuss the recent development of a novel sparse regression technique that aims to accurately decompose process variation into two different components: (1) spatially correlated variation, and (2) ...
U-226: Linux Kernel SFC Driver TCP MSS Option Handling Denial...
Broader source: Energy.gov (indexed) [DOE]
Vulnerability PLATFORM: Linux Kernel 3.2.x ABSTRACT: The Linux kernel is prone to a remote denial-of-service vulnerability. reference LINKS: Secunia Advisory SA50081 Bugtraq ID:...
T-583: Linux Kernel OSF Partition Table Buffer Overflow Lets Local Users Obtain Information
Broader source: Energy.gov [DOE]
A local user can create a storage device with specially crafted OSF partition tables. When the kernel automatically evaluates the partition tables, a buffer overflow may occur and data from kernel heap space may leak to user-space.
Kollegala, Revathi
2012-07-16T23:59:59.000Z
of wavelet functions as kernels with Support Vector Data Description for target detection in hyperspectral images. Specifically, it proposes the Adaptive Wavelet Kernel Support Vector Data Description (AWK-SVDD) that learns the optimal wavelet function...
A comparative study of spline regression
Nougues, Arnaud
1980-01-01T23:59:59.000Z
POIRIER'S METHOD FOR COMPUTING LEAST SQUARES SPLINES 4. 2 THE RESTRICTED LEAST SQUARES APPROACH 4. 3 LEAST SQUARES SPLINES USING THE B 15 REPRESENTATION 21 4. 4 LEAST SQUARES SPLINES USING THE TRUNCATED POWER BASIS REPRESENTATION 24 4. 5 A...: RATIONALE FOR POIRIER'S METHOD APPENDIX B: A LEAST SQUARES CUBIC SPLINE PROGRAM BASED ON POIRIER'S METHOD ~Pa e 52 55 APPENDIX C: COMPUTATION OF B IN THE RESTRICTED LEAST SQUARES METHOD FOR CUBIC SPLINES WITH A CONTINUOUS SECOND DERIVATIVE APPENDIX...
Iterative gradient descent approach to multiple regression with fuzzy data
Bargiela, Andrzej
to multiple regression and lay foundation for a further generalisation to multiple non-linear regression dictates adoption of a more general viewpoint, regression variables are given as non-numerical entities of the parameters of the regression model have been derived only for the case of a simple linear regression, i
Math 261A -Spring 2012 M. Bremer Multiple Linear Regression
Keinan, Alon
called non-linear regression models or polynomial regression models, as the regression curveMath 261A - Spring 2012 M. Bremer Multiple Linear Regression So far, we have seen the concept of simple linear regression where a single predictor variable X was used to model the response variable Y
The Equivalence between Row and Column Linear Regression
Tresp, Volker
The Equivalence between Row and Column Linear Regression: A Surprising Feature of Linear Regression and Communications 81730 M¨unchen, Germany Abstract The rows of the design matrix in linear regression are the inputs of the training data set. Normal regression is row regression: the goal is to predict a training target from
Geometry, noncommutative algebra and representations
Gordon, Iain
and Deformations 4 Representation Theory 2 Iain Gordon Geometry, noncommutative algebra and representations: analysis, algebra, geometry, number theory (to name four!) 4 Iain Gordon Geometry, noncommutative algebra is a finite field. 6 Iain Gordon Geometry, noncommutative algebra and representations Geometry and Commutative
A Kalman-Particle Kernel Filter and its Application to Terrain Navigation
Del Moral , Pierre
A Kalman-Particle Kernel Filter and its Application to Terrain Navigation Dinh-Tuan Pham.musso@onera.fr Abstract A new nonlinear filter, the Kalman- Particle Kernel Filter (KPKF) is proposed. Compared. Keywords: Kalman filter, kernel density estimator, regularized particle filter, Inertial navigation System
Sparse Kernel Orthonormalized PLS for feature extraction in large data sets
Sparse Kernel Orthonormalized PLS for feature extraction in large data sets Anonymous Author a novel multivariate analysis method for large scale problems. Our scheme is based on a novel kernel strong expressive power even with rather few features, is clearly outperforming the ordinary kernel PLS
Boyer, Edmond
). The case of stable random elements was also investigated (see for instance Li, Linde (2004), Aurzada, Lifshits, Linde (2009)). Another issue is related to the norm. Indeed in infinite dimensional spaces, norms) or Li and Linde (1993) for instance. A classical example stems from the situation where PX-x0 PX where
Distributed Multivariate Regression Using Wavelet-based Collective Data Mining.
Kargupta, Hilol
Distributed Multivariate Regression Using Wavelet-based Collective Data Mining. Daryl E a method for distributed multivariate regression using wavelet- based Collective Data Mining (CDM employed in parametric multivariate regression to provide an effective data mining technique for use
Identifying and removing sources of imprecision in polynomial regression
Brauner, Neima
Identifying and removing sources of imprecision in polynomial regression Neima Braunera and removal of imprecision in polynomial regression, originating from random errors (noise) in the independent.V. Keywords: Regression; Polynomial; Precision; Noise; Collinearity 1. Introduction Mathematical modeling
IRT Goodness-of-Fit Using Approaches from Logistic Regression
Mair, Patrick; Reise, Steven P.; Bentler, Peter M.
2008-01-01T23:59:59.000Z
connect IRT and logistic regression, we represent the matrixApproaches from Logistic Regression us write the log-oddsApproaches from Logistic Regression Cox DR, Snell EJ (1989).
Computationally Efficient Gaussian Process Changepoint Detection and Regression
Computationally Efficient Gaussian Process Changepoint Detection and Regression by Robert Conlin Changepoint Detection and Regression by Robert Conlin Grande Submitted to the Department of Aeronautics in Aerospace Engineering Abstract Most existing GP regression algorithms assume a single generative model
Regression: What's it all about?1 Andrew Gelman2
Gelman, Andrew
Regression: What's it all about?1 Andrew Gelman2 5 Jan 2015 Regression plays three different roles in applied statistics: 1. A specification the different faces of regression modeling after being asked to review the new book
IRT Goodness-of-Fit Using Approaches from Logistic Regression
Patrick Mair; Steven P. Reise; Peter M. Bentler
2011-01-01T23:59:59.000Z
connect IRT and logistic regression, we represent the matrixApproaches from Logistic Regression us write the log-oddsApproaches from Logistic Regression Cox DR, Snell EJ (1989).
Reproducing kernel element method Part III: Generalized enrichment and applications
Li, Shaofan
Reproducing kernel element method Part III: Generalized enrichment and applications Hongsheng Lu enrichment is proposed to construct the global partition polynomials or to enrich global partition polynomial. This is accomplished by either multiplying enrichment functions with the original global partition poly- nomials
Improving the Energy Efficiency of the MANTIS Kernel
Sreenan, Cormac J.
Improving the Energy Efficiency of the MANTIS Kernel Cormac Duffy1 , Utz Roedig2 , John Herbert1. The event-based TinyOS is more energy efficient than the multi-threaded MANTIS system. However, MANTIS, timeliness can be traded for energy efficiency by choosing the appropriate operating system. In this paper we
Optical transformation from chirplet to fractional Fourier transformation kernel
Hong-yi Fan; Li-yun Hu
2009-02-11T23:59:59.000Z
We find a new integration transformation which can convert a chirplet function to fractional Fourier transformation kernel, this new transformation is invertible and obeys Parseval theorem. Under this transformation a new relationship between a phase space function and its Weyl-Wigner quantum correspondence operator is revealed.
Measurement Denoising Using Kernel Adaptive Filters in the Smart Grid
Qiu, Robert Caiming
Measurement Denoising Using Kernel Adaptive Filters in the Smart Grid Zhe Chen and Robert C. Qiu@ieee.org, rqiu@tntech.edu Abstract--State estimation plays an important role in the smart grid. Conventionally, noisy measurements are directly used for state estimation. Today, in the context of the smart grid
Kernel spectral clustering for predicting maintenance of industrial machines
status, but the operations are labor inten- sive and prone to human errors. Condition-based maintenance monitoring of machine parts leads to reliable and accurate lifetime predictions, and maintenance operationsKernel spectral clustering for predicting maintenance of industrial machines Rocco Langone1, Carlos
A Kernel Method for Market Clearing Sebastien Lahaie
Sandholm, Tuomas W.
linear prices do not suffice. We first present a procedure that, given a sample of values and costs nonlinear clearing prices with much less than full revelation of values and costs. When the kernel function that, given a sample of values and costs for a set of bundles, computes nonlinear clearing prices using
ASYMPTOTIC PROPERTIES OF THE HEAT KERNEL ON CONIC MANIFOLDS
Loya, Paul
ASYMPTOTIC PROPERTIES OF THE HEAT KERNEL ON CONIC MANIFOLDS PAUL LOYA Abstract. We derive Foundation Fellowship. 1 #12; 2 PAUL LOYA Trace expansions of cone operators has a long history stemming from on conic manifolds; see for instance, Callias [5], Cheeger [7], Chou [9], BrË?uning--Seeley [3], Br
Kernel Methods for Melanoma Recognition Elisabetta LA TORREa,1
Caputo, Barbara
Kernel Methods for Melanoma Recognition Elisabetta LA TORREa,1 , Tatiana TOMMASIa , Barbara CAPUTOb for computer assisted diagnosis of melanomas. The first is the support vector machines algorithm, a state a sophisticated segmentation technique and a set of features especially designed for melanoma recognition. To our
Melanoma Recognition Using Representative and Discriminative Kernel Classifiers
Caputo, Barbara
Melanoma Recognition Using Representative and Discriminative Kernel Classifiers Tatiana Tommasi1 caputo@nada.kth.se Abstract. Malignant melanoma is the most deadly form of skin lesion. Early diagnosis these algorithms against the (to our knowledge) state-of-the-art method on melanoma recognition, exploring how
PERI - Auto-tuning Memory Intensive Kernels for Multicore
Bailey, David H; Williams, Samuel; Datta, Kaushik; Carter, Jonathan; Oliker, Leonid; Shalf, John; Yelick, Katherine; Bailey, David H
2008-06-24T23:59:59.000Z
We present an auto-tuning approach to optimize application performance on emerging multicore architectures. The methodology extends the idea of search-based performance optimizations, popular in linear algebra and FFT libraries, to application-specific computational kernels. Our work applies this strategy to Sparse Matrix Vector Multiplication (SpMV), the explicit heat equation PDE on a regular grid (Stencil), and a lattice Boltzmann application (LBMHD). We explore one of the broadest sets of multicore architectures in the HPC literature, including the Intel Xeon Clovertown, AMD Opteron Barcelona, Sun Victoria Falls, and the Sony-Toshiba-IBM (STI) Cell. Rather than hand-tuning each kernel for each system, we develop a code generator for each kernel that allows us to identify a highly optimized version for each platform, while amortizing the human programming effort. Results show that our auto-tuned kernel applications often achieve a better than 4X improvement compared with the original code. Additionally, we analyze a Roofline performance model for each platform to reveal hardware bottlenecks and software challenges for future multicore systems and applications.
A Forced Sampled Execution Approach to Kernel Rootkit Identification
Chiueh, Tzi-cker
}@symantec.com Abstract. Kernel rootkits are considered one of the most dangerous forms of malware because they reside a forced sampled execution approach to traverse the driver's control flow graph. Through a comprehensive, 11, 4] is a piece of binary code that a computer in- truder, after breaking into a machine, installs
A comparative study of spline regression
Nougues, Arnaud
1980-01-01T23:59:59.000Z
ta IC 33 6. 00 3. 00 '!. 00 5 00 6. DO 1, 00 Oiaa 9. 00 10. 00 11 00 Il. tla IC. Ita I'1. 00 NUBBER BF BRERKPOiNOB CI al 01 0'3 3. 00 a. OD 5, 00 6. DD 1. 00 Oi tla . OD 10. 0 ~ I I, 30 13, OD 13. 00 10. 00 NUNBFR 0, 33ERKPBiNTB Figure 5... POIRIER'S METHOD FOR COMPUTING LEAST SQUARES SPLINES 4. 2 THE RESTRICTED LEAST SQUARES APPROACH 4. 3 LEAST SQUARES SPLINES USING THE B 15 REPRESENTATION 21 4. 4 LEAST SQUARES SPLINES USING THE TRUNCATED POWER BASIS REPRESENTATION 24 4. 5 A...
Ensemble Kalman Filtering with Shrinkage Regression Techniques
Eidsvik, Jo
Ensemble Kalman Filtering with Shrinkage Regression Techniques Jon Sćtrom & Henning Omre, Norwegian University of Science and Technology; Summary The classical Ensemble Kalman Filter (EnKF) is known;Introduction The Ensemble Kalman Filter (EnKF) is a Bayesian data assimilation method that in recent years has
Calibration via Regression Dean P. Foster
Kakade, Sham M.
Calibration via Regression Dean P. Foster Statistics Department University of Pennsylvania Email-- In the online prediction setting, the concept of calibration entails having the empirical (conditional hard to compare with each other. This paper shows how to get an approximate form of calibration out
Developmental Regression in Children with Down Syndrome
Bernad Ripoll, Susana
2011-05-18T23:59:59.000Z
This study presents in detail the data on a group of 20 participants, female and male, from 2 to 12 years old with Down syndrome (DS) who experienced developmental regression. This study took place at the Down syndrome clinic at Kennedy Krieger...
Geodesic Regression on Riemannian Manifolds P. Thomas Fletcher
Boyer, Edmond
- ing multiple linear regression in Rn . Here we are interested in the relationship between a non that one could choose, and it provides a direct generalization of linear regression to the manifold setting regression model is linear regression, due to its simplicity, ease of interpretation, and ability to model
Morse-Smale Regression Samuel Gerber, University of Utah
approximated by a linear model. This approach yields regression models that are amenable to interpretation capabilities of non-parametric methods. A classical 1 #12;approach to partition-based regression are regression a piece- wise constant model, treed regression (Alexander and Grimshaw, 1996) proposed linear models
Probability Density Function Estimation Using Orthogonal Forward Regression
Chen, Sheng
Probability Density Function Estimation Using Orthogonal Forward Regression S. Chen, X. Hong and C estimation is formulated as a regression problem and the orthogonal forward regression tech- nique is adopted procedure. Two examples are used to demonstrate the ability of this regression- based approach
Quantile Regression for Correlated Observations , Lee-Jen Wei2
Wolfe, Patrick J.
Quantile Regression for Correlated Observations Li Chen1 , Lee-Jen Wei2 , and Michael I. Parzen 1 60637 Abstract In this paper, we consider the problem of regression analysis for data which consist the standard mean regression, we regress various percentiles of each marginal response variable over its
Multiple Regression Introduction to Statistics Using R (Psychology 9041B)
Gribble, Paul
Multiple Regression Introduction to Statistics Using R (Psychology 9041B) Paul Gribble Winter, 2015 1 Correlation, Regression & Multiple Regression 1.1 Bivariate correlation The Pearson product*r*(N-2)) / (1-(r*r)) > pobs Fobs 1 #12;Intro Stats R 1 Correlation, Regression
Local Regression with Meaningful Parameters Rafael A. Irizarry
Irizarry, Rafael A.
Local Regression with Meaningful Parameters Rafael A. Irizarry Abstract Local regression or loess squares. In this paper we will present a version of local regression that fits more general parametric, and financial data are included. KEY WORDS: Local Regression, Harmonic Model, Meaningful Parameters, Sound
JOURNAL OF LATEX CLASS FILES 1 Robust Logistic Regression
Hero, Alfred O.
JOURNAL OF LATEX CLASS FILES 1 Robust Logistic Regression with Bounded Data Uncertainties Patrick L a formulation of robust logistic regression under bounded data uncertainties. The robust estimates are obtained compare the results of 1-Logistic Regression against 1-Robust Logistic Regression on a real gene
Censored Regression Trend Analyses for Ambient Water Quality Data
Smyth, Gordon K.
Censored Regression Trend Analyses for Ambient Water Quality Data Gordon K. Smyth 1 , Melanie Cox 2 regression; logistic distribution; regression splines; seasonal trends. 1 Introduction Water is a very. A censored regression strat- egy is used to accommodate arbitrary detection limits for the indicator
Ridge Regression Estimation Approach to Measurement Error Model
Shalabh
Ridge Regression Estimation Approach to Measurement Error Model A.K.Md. Ehsanes Saleh Carleton of the regression parameters is ill conditioned. We consider the Hoerl and Kennard type (1970) ridge regression (RR) modifications of the five quasi- empirical Bayes estimators of the regression parameters of a measurement error
Deepest Regression in Analytical Chemistry P. J. Rousseeuw,a
Van Aelst, Stefan
Deepest Regression in Analytical Chemistry P. J. Rousseeuw,a , S. Van Aelst,a , B. Rambalib , and J Recently the concept of regression depth has been introduced [1]. The deepest regression (DR) is a method for linear regression which is defined as the fit with the best depth relative to the data. In this paper we
Isotonic Regression for Multiple Independent Variables Quentin F. Stout
Stout, Quentin F.
Isotonic Regression for Multiple Independent Variables Quentin F. Stout Computer Science for determining isotonic regressions for weighted data at a set of points P in multidimensional space isotonic regression with unweighted data. L isotonic regression is not unique, and algorithms are given
REGRESSION BY SELECTING APPROPRIATE 7ROJD$\\GQDQG+$OWD\\*venir
Güvenir, H. Altay
REGRESSION BY SELECTING APPROPRIATE FEATURE(S) 7ROJD$\\GŐQDQG+$OWD\\*üvenir Department of Computer methods, called Regression by Selecting Best Feature Projection (RSBFP) and Regression by Selecting Best linear regression line on each continuous feature. In the case of categorical features, exactly one
Positive curvature property for some hypoelliptic heat kernels
Qian, Bin
2010-01-01T23:59:59.000Z
In this note, we look at some hypoelliptic operators arising from nilpotent rank 2 Lie algebras. In particular, we concentrate on the diffusion generated by three Brownian motions and their three L\\'evy areas, which is the simplest extension of the Laplacian on the Heisenberg group $\\mathbb{H}$. In order to study contraction properties of the heat kernel, we show that, as in the case of the Heisenberg group, the restriction of the sub-Laplace operator acting on radial functions (which are defined in some precise way in the core of the paper) satisfies a non negative Ricci curvature condition (more precisely a $CD(0, \\infty)$ inequality), whereas the operator itself does not satisfy any $CD(r,\\infty)$ inequality. From this we may deduce some useful, sharp gradient bounds for the associated heat kernel.
Memory Kernel in the Expertise of Chess Players
Schaigorodsky, Ana L; Billoni, Orlando V
2015-01-01T23:59:59.000Z
In this work we investigate a mechanism for the emergence of long-range time correlations observed in a chronologically ordered database of chess games. We analyze a modified Yule-Simon preferential growth process proposed by Cattuto et al., which includes memory effects by means of a probabilistic kernel. According to the Hurst exponent of different constructed time series from the record of games, artificially generated databases from the model exhibit similar long-range correlations. In addition, the inter-event time frequency distribution is well reproduced by the model for realistic parameter values. In particular, we find the inter-event time distribution properties to be correlated with the expertise of the chess players through the memory kernel extension. Our work provides new information about the strategies implemented by players with different levels of expertise, showing an interesting example of how popularities and long-range correlations build together during a collective learning process.
A Gibbs Sampler for Multivariate Linear Regression
Mantz, Adam B
2015-01-01T23:59:59.000Z
Kelly (2007, hereafter K07) described an efficient algorithm, using Gibbs sampling, for performing linear regression in the fairly general case where non-zero measurement errors exist for both the covariates and response variables, where these measurements may be correlated (for the same data point), where the response variable is affected by intrinsic scatter in addition to measurement error, and where the prior distribution of covariates is modeled by a flexible mixture of Gaussians rather than assumed to be uniform. Here I extend the K07 algorithm in two ways. First, the procedure is generalized to the case of multiple response variables. Second, I describe how to model the prior distribution of covariates using a Dirichlet process, which can be thought of as a Gaussian mixture where the number of mixture components is learned from the data. I present an example of multivariate regression using the extended algorithm, namely fitting scaling relations of the gas mass, temperature, and luminosity of dynamica...
Symbolic Representation of Neural Networks
Liu, Huan
Draft Symbolic Representation of Neural Networks Rudy Setiono and Huan Liu Department,liuhg@iscs.nus.sg Abstract Although backpropagation neural networks generally predict better than decision trees do is needed by human experts. This work drives a sym bolic representation for neural networks to make
U-080: Linux Kernel XFS Heap Overflow May Let Remote Users Execute Arbitrary Code
Broader source: Energy.gov [DOE]
A vulnerability was reported in the Linux Kernel. A remote user can cause arbitrary code to be executed on the target user's system.
A high-order fast method for computing convolution integral with smooth kernel
Qiang, Ji
2010-01-01T23:59:59.000Z
of the convolution integral. The downside of these evenfor computing convolution integral with smooth kernel Jicalculate convolution integral with smooth non-periodic
Weighted Bergman Kernel Functions and the Lu Qi-keng Problem
Jacobson, Robert Lawrence
2012-07-16T23:59:59.000Z
Weighted Bergman Kernel Functions and the Lu Qi-keng Problem. (May 2012) Robert Lawrence Jacobson, B.S, Southern Adventist University Chair of Advisory Committee: Dr. Harold P. Boas The classical Lu Qi-keng Conjecture asks whether the Bergman kernel... meromorphic kernel with a nite number of zeros on the domain. For kernels associated to meromorphic iv functions with an arbitrary number of zeros on the domain, we obtain a weighted ver- sion of the classical Ramadanov?s Theorem which says that for a...
T-653: Linux Kernel sigqueueinfo() Process Lets Local Users Send Spoofed Signals
Broader source: Energy.gov [DOE]
A vulnerability was reported in the Linux Kernel. A local user can send spoofed signals to other processes in certain cases.
Nonlinear regression analysis of field emission data
Barry, Scott Wilson
1992-01-01T23:59:59.000Z
for the zirconium/tungsten cathode data. . Regressed enhancement fa. ctors using the integral model(solid line) and approximate model(dashed line) over a range of fixed work functions for the zirconium/tungsten cathode data. 70 Integral model(solid line...) and linear( dashed line) fitting curves for the zirconium/tungsten cathode data, . 71 33 Integral model(solid line) and linear(dashed line) fitting curves for the zirconium/tungsten cathode data, excluding the last three suspect data points. 72 CHAPTER I...
Honest Confidence Intervals for the Error Variance in Stepwise Regression
Stine, Robert A.
Honest Confidence Intervals for the Error Variance in Stepwise Regression Dean P. Foster and Robert alternatives are used. These simpler algorithms (e.g., forward or backward stepwise regression) obtain
Efficient Learning and Feature Selection in High Dimensional Regression
Ting, Jo-Anne; D'Souza, Aaron; Vijayakumar, Sethu; Schaal, Stefan
2010-01-01T23:59:59.000Z
We present a novel algorithm for efficient learning and feature selection in high-dimensional regression problems. We arrive at this model through a modification of the standard regression model, enabling us to derive a probabilistic version...
Testing regression models with residuals as data by Xia Hua.
Hua, Xia, Ph. D. Massachusetts Institute of Technology
2010-01-01T23:59:59.000Z
Abstract In polynomial regression ... . In this thesis, I developed a residual based test, the turning point test for residuals, which tests the hypothesis that the kth order polynomial regression holds with ... while the ...
Local Dimensionality Reduction for Non-Parametric Regression
Hoffmann, Heiko; Schaal, Stefan; Vijayakumar, Sethu
2009-01-01T23:59:59.000Z
Locally-weighted regression is a computationally-efficient technique for non-linear regression. However, for high-dimensional data, this technique becomes numerically brittle and computationally too expensive if many ...
Regression Given input data (features), predict value of a
Giger, Christine
· The complete graph Non-linear regression #12;· Need to fit non-linear functions · example: polynomials Non-linear of the inputs · After applying z(x), we fit a plane in 3D-space Non-linear regression y = 0 + 1x + 2x2 z(x) = 1 x x2 #12;x y x2 regression from x to y, non-linear lifting from x to z=[x x2] regression from [x x2
Adaptive sparse polynomial chaos expansion based on least angle regression
Blatman, Geraud, E-mail: geraud.blatman@edf.f [Clermont Universite, IFMA, EA 3867, Laboratoire de Mecanique et Ingenieries, BP 10448, F-63000 Clermont-Ferrand (France); EDF R and D, Departement Materiaux et Mecanique des Composants, Site des Renardieres, 77250 Moret-sur-Loing cedex (France); Sudret, Bruno [Clermont Universite, IFMA, EA 3867, Laboratoire de Mecanique et Ingenieries, BP 10448, F-63000 Clermont-Ferrand (France); Phimeca Engineering, Centre d'Affaires du Zenith, 34 rue de Sarlieve, F-63800 Cournon d'Auvergne (France)
2011-03-20T23:59:59.000Z
Polynomial chaos (PC) expansions are used in stochastic finite element analysis to represent the random model response by a set of coefficients in a suitable (so-called polynomial chaos) basis. The number of terms to be computed grows dramatically with the size of the input random vector, which makes the computational cost of classical solution schemes (may it be intrusive (i.e. of Galerkin type) or non intrusive) unaffordable when the deterministic finite element model is expensive to evaluate. To address such problems, the paper describes a non intrusive method that builds a sparse PC expansion. First, an original strategy for truncating the PC expansions, based on hyperbolic index sets, is proposed. Then an adaptive algorithm based on least angle regression (LAR) is devised for automatically detecting the significant coefficients of the PC expansion. Beside the sparsity of the basis, the experimental design used at each step of the algorithm is systematically complemented in order to avoid the overfitting phenomenon. The accuracy of the PC metamodel is checked using an estimate inspired by statistical learning theory, namely the corrected leave-one-out error. As a consequence, a rather small number of PC terms are eventually retained (sparse representation), which may be obtained at a reduced computational cost compared to the classical 'full' PC approximation. The convergence of the algorithm is shown on an analytical function. Then the method is illustrated on three stochastic finite element problems. The first model features 10 input random variables, whereas the two others involve an input random field, which is discretized into 38 and 30 - 500 random variables, respectively.
Kernel-based distance metric learning for microarray data classification
Xiong, Huilin; Chen, Xue-wen
2006-06-01T23:59:59.000Z
the test data. We only consider Gaussian kernel function in the proposed and SVM algorithms. 1. ALL-AML Leukemia Data: This data set, taken from the website [17], contains 72 samples of human acute leuke- mia. 47 samples belong to acute lymphoblastic... lymphoblastic leukemia (ALL), 20 of them to mixed line- age leukemia (MLL), a subset of human acute leukemia with a chromosomal translocation, and 28 of the samples are acute myelogenous leukemia (AML). Each sample gives the expression levels of 12582 genes...
Kernelization and Enumeration: New Approaches to Solving Hard Problems
Meng, Jie
2011-08-08T23:59:59.000Z
ABSTRACT Kernelization and Enumeration: New Approaches to Solving Hard Problems. (May 2010) Jie Meng, B.S.; M.S., Fudan University Chair of Advisory Committee: Dr. Jianer Chen NP-Hardness is a well-known theory to identify the hardness of computational... veri able if there is an algorithm A such that: Given an instance s and a proof for s, A could verify whether s is a Yes-instance of P ; The runtime of A is bounded by a polynomial in the size of s; For example, the k-path problem asks...
A Hybrid GP Approach for Numerically Robust Symbolic Regression
Fernandez, Thomas
expressions encoded in tree structures to perform symbolic regression. A non-linear optimization method very common technique is linear regression, in which the model is a linear combina- tion of given base are polynomials (polynomial regression) or trigonometric poly- nomials (e.g. Fourier series). For general linearly
Non-linear regression models for Approximate Bayesian Computation
Robert, Christian P.
Non-linear regression models for Approximate Bayesian Computation (ABC) Michael Blum Olivier ABC #12;Blum and OF (2009) suggest the use of non-linear conditional heteroscedastic regression models) Linear regression-based ABC can sometimes be improved #12;abc of ABC Using stochastic simulations
The SROV program for data analysis and regression model identification
Brauner, Neima
) regression models comprised of linear combination of independent variables and their non-linear functions. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Stepwise regression; Colinearity; Non have been intro- duced for selection of the optimal model in linear regression (for detailed reviews
Submitted to the Annals of Statistics FUNCTIONAL ADDITIVE REGRESSION
Radchenko, Peter
extends beyond the standard linear regression setting to fit general non-linear additive models. We extending the classical functional regression model. [25] proposed an index model to implement a non-linear extends the usual linear regression model involving a functional predictor, X(t), and a scalar response, Y
Moment Based Dimension Reduction for Multivariate Response Regression
Bura, Efstathia
Moment Based Dimension Reduction for Multivariate Response Regression Xiangrong Yin Efstathia Bura January 20, 2005 Abstract Dimension reduction aims to reduce the complexity of a regression without re- quiring a pre-specified model. In the case of multivariate response regressions, covariance
Ridge Regression Estimation for Survey Samples Mingue Park
Yang, Min
Ridge Regression Estimation for Survey Samples Mingue Park and Min Yang Korea University and University of Missouri Abstract A procedure for constructing a vector of regression weights is considered. Under the re- gression superpopulation model, the ridge regression estimator that has minimum model mean
A note on modelling cross correlations: hyperbolic secant regression
Smyth, Gordon K.
A note on modelling cross correlations: hyperbolic secant regression Gordon K. Smyth Department of size one. This al- lows regression type modelling of the correlation without unnecessary loss ~ is given in Section 2 that has useful performance down to samples of size one. This allows regression type
Nonparametric Regression in Exponential Families Lawrence D. Brown1
Zhou, Harrison Huibin
Nonparametric Regression in Exponential Families Lawrence D. Brown1 , T. Tony Cai2 and Harrison H. Zhou3 University of Pennsylvania and Yale University Abstract Most results in nonparametric regression we consider nonparametric regression in exponential families which include, for example, Poisson
An Efficient Algorithm for REML in Heteroscedastic Regression
Smyth, Gordon K.
An Efficient Algorithm for REML in Heteroscedastic Regression Gordon K. Smyth Walter and Eliza Hall of covariates relevant for predicting the variance and and are vectors of regression coefficients. The model). Heteroscedastic regression models have an extensive literature going back to Park (1966), Rutemiller & Bowers
NONPARAMETRIC REGRESSION ESTIMATION OF HOUSEHOLD VMT Young-Jun Kweon
Kockelman, Kara M.
NONPARAMETRIC REGRESSION ESTIMATION OF HOUSEHOLD VMT Young-Jun Kweon (Corresponding Author Transportation Survey. The results are density functions and regression surfaces for VMT, in relation regressions perform better than their ordinary-least squares counterparts in many ways: they permit full
Practical Regression and Anova using R Julian J. Faraway
Practical Regression and Anova using R Julian J. Faraway July 2002 #12;1 Copyright c 1999, 2000 There are many books on regression and analysis of variance. These books expect different levels of pre is on the practice of regression and analysis of variance. The objective is to learn what methods are available
Conditional Regression Forests for Human Pose Estimation Pushmeet Kohli
Kohli, Pushmeet
Conditional Regression Forests for Human Pose Estimation Min Sun Pushmeet Kohli Jamie Shotton estimation from depth images. The conditional regression model proposed in the paper is general and can body joint prediction as a regression problem which avoids intermediate body part classification
AUTOMATED REGRESSION TESTING AND VERIFICATION OF COMPLEX CODE
Roychoudhury, Abhik
AUTOMATED REGRESSION TESTING AND VERIFICATION OF COMPLEX CODE CHANGES DOCTORAL THESIS MARCEL B¨OHME NATIONAL UNIVERSITY OF SINGAPORE 2014 #12;AUTOMATED REGRESSION TESTING AND VERIFICATION OF COMPLEX CODE : Automated Regression Testing and Verification of Complex Code Changes Abstract How can we check software
Analysis of Some Methods for Reduced Rank Gaussian Process Regression
Analysis of Some Methods for Reduced Rank Gaussian Process Regression Joaquin Qui~nonero-Candela1 there is strong motivation for using Gaussian Pro- cesses (GPs) due to their excellent performance in regression-effective ap- proximations to GPs, both for classification and for regression. In this paper we analyze one
Nonparametric Regression in Exponential Families Lawrence D. Brown1
Brown, Lawrence D.
Nonparametric Regression in Exponential Families Lawrence D. Brown1 , T. Tony Cai1 and Harrison H. Zhou2 University of Pennsylvania and Yale University Abstract Most results in nonparametric regression we consider nonparametric regression in exponential families with the main focus on the natural
Robust Regression Analysis: Some Popular Statistical Package Options
Masci, Frank
1 Robust Regression Analysis: Some Popular Statistical Package Options By Robert A. Yaffee Meyer of SAS, Institute for their gracious assistance. Robust regression analysis provides an alternative to a least squares regression model when fundamental assumptions are unfulfilled by the nature
ccsd-00085858,version1-16Jul2006 SUPERCONNECTION AND FAMILY BERGMAN KERNELS
Boyer, Edmond
ccsd-00085858,version1-16Jul2006 SUPERCONNECTION AND FAMILY BERGMAN KERNELS XIAONAN MA AND WEIPING is to use the superconnection as in the local family index theorem. Superconnexion et noyaux de Bergman en;SUPERCONNECTION AND FAMILY BERGMAN KERNELS 3 Let T 2 (T RW) TRX be the tensor defined in the following way
Random Features for Large-Scale Kernel Machines Intel Research Seattle
Kim, Tae-Kyun
Random Features for Large-Scale Kernel Machines Ali Rahimi Intel Research Seattle Seattle, WA 98105 products of the transformed data are approximately equal to those in the feature space of a user specified on their ability to approximate various radial basis kernels, and show that in large-scale classification
Hash-SVM: Scalable Kernel Machines for Large-Scale Visual Classification , Shih-Fu Chang
Chang, Shih-Fu
Hash-SVM: Scalable Kernel Machines for Large-Scale Visual Classification Yadong Mu , Gang Hua , Wei the efficiency of non-linear kernel SVM in very large scale visual classification prob- lems. Our key idea be transformed into solving a linear SVM over the hash bits. The proposed Hash-SVM enjoys dramatic storage cost
JKernelMachines: A Simple Framework for Kernel Machines David Picard PICARD@ENSEA.FR
Paris-Sud XI, Université de
. Description of the Library The backbone of the library is the definition of data types and kernels. In order to use any type of input space, the library makes heavy use o, France Editor: Abstract JKernelMachines is a Java library for learning with kernels. It is primarily
An improved ECG-Derived Respiration Method using Kernel Principal Component Analysis
) of heart beats generates well-performing ECG- derived respiratory signals (EDR). This study aims at im- proving the performance of EDR signals using kernel PCA (kPCA). Kernel PCA is a generalization of PCA and kPCA is eval- uated by comparing the EDR signals to the reference res- piratory signal. Correlation
Application of Kernel Principal Component Analysis for Single Lead ECG-Derived Respiration
signal from ECGs. In this study, an improved ECG-derived respiration (EDR) algorithm based on kernel PCA function (RBF) kernel performs the best when deriving EDR signals. Further improvement is carried outPCA is assessed by comparing the EDR signals to a reference respiratory signal, using the correlation
HW Componentizing Kernel: A New Approach to address the Mega Complexity of Future Automotive CPS
Rajkumar, Ragunathan "Raj"
HW Componentizing Kernel: A New Approach to address the Mega Complexity of Future Automotive CPS of CPS (Cyber Physical System). However, current software development process in the automotive industry automotive software devel- opment process in the perspective of CPS and proposes a new kernel-based approach
Comparing Kernels For Predicting Protein Binding Sites From Amino Acid Sequence
Honavar, Vasant
in on all three tasks, with the substitution kernel based on amino acid substitution matrices that take into account structural or evolutionary conservation or physicochemical properties of amino acids yields modestComparing Kernels For Predicting Protein Binding Sites From Amino Acid Sequence Feihong Wu1
Villarreal Lozoya, Jose Emilio
2009-05-15T23:59:59.000Z
Pecan kernels from six cultivars were analyzed for phenolic content and antioxidant properties. In addition, kernels from two cultivars were irradiated with 0, 1.5 and 3.0 kGy using E-Beam irradiation and stored in accelerated conditions (40 Â...
Kernel de Tiempo Real para Control de Procesos Oscar Miranda Gomez, Pedro Mejia Alvarez
Mejia-Alvarez, Pedro
´afico de simulaci´on de tareas de tiempo real. El Kernel de Tiempo Real es peque~no (de tama~no 20.5Kb) y procesador y un peque~no sistema operativo empotrado el cual es capaz de controlar todo el hardware de manera de tiempo real el principal compo- nente lo constituye el sistema operativo o Kernel. Un sistema
Villarreal Lozoya, Jose Emilio
2009-05-15T23:59:59.000Z
Pecan kernels from six cultivars were analyzed for phenolic content and antioxidant properties. In addition, kernels from two cultivars were irradiated with 0, 1.5 and 3.0 kGy using E-Beam irradiation and stored in accelerated conditions (40 Â°C...
Supplemental Material for "Classifying Video with Kernel Dynamic Antoni B. Chan and Nuno Vasconcelos
Vasconcelos, Nuno M.
Supplemental Material for "Classifying Video with Kernel Dynamic Textures" Antoni B. Chan and Nuno Vasconcelos SVCL-TR 2007/03 April 2007 #12;#12;Supplemental Material for "Classifying Video with Kernel This is the supplemental material for "Classifying Video with Dynamic Textures" [1]. It contains information about
Obradovic, Zoran
of accuracy for both linear and non-linear regression models. The obtained experimental results suggest impact on accuracy of an auto-regression model. For non-linear phenomena, learning algorithms that model grid. The proposed method combines linear or non-linear non-spatial and non- temporal regression models
Quantum Field Theory and Representation Theory
Woit, Peter
Quantum Field Theory and Representation Theory Peter Woit woit@math.columbia.edu Department of Mathematics Columbia University Quantum Field Theory and Representation Theory p.1 #12;Outline of the talk · Quantum Mechanics and Representation Theory: Some History Quantum Field Theory and Representation Theory
Towards Optimal and Expressive Kernelization for d-Hitting Set
van Bevern, René
2011-01-01T23:59:59.000Z
A sunflower in a hypergraph is a set of hyperedges pairwise intersecting in exactly the same vertex set. Sunflowers are a useful tool in polynomial-time data reduction for problems formalizable as d-Hitting Set, the problem of covering all hyperedges (of cardinality at most d) of a hypergraph by at most k vertices. Additionally, in fault diagnosis, sunflowers yield concise explanations for "highly defective structures". We provide a linear-time algorithm that, by finding sunflowers, transforms an instance of d-Hitting Set into an equivalent instance comprising at most O(k^d) hyperedges and vertices. In terms of parameterized complexity theory, we show a problem kernel with asymptotically optimal size (unless coNP in NP/poly). We show that the number of vertices can be reduced to O(k^(d-1)) with additional processing in O(k^(1.5d)) time---nontrivially combining the sunflower technique with a known problem kernel that uses crown reductions.
Jumping Neptune Can Explain the Kuiper Belt Kernel
Nesvorny, David
2015-01-01T23:59:59.000Z
The Kuiper belt is a population of icy bodies beyond the orbit of Neptune. A particularly puzzling and up-to-now unexplained feature of the Kuiper belt is the so-called `kernel', a concentration of orbits with semimajor axes a~44 AU, eccentricities e~0.05, and inclinations ibelt kernel can be explained if Neptune's otherwise smooth migration was interrupted by a discontinuous change of Neptune's semimajor axis when Neptune reached ~28 AU. Before the discontinuity happened, planetesimals located at ~40 AU were swept into Neptune's 2:1 resonance, and were carried with the migrating resonance outwards. The 2:1 resonance was at ~44 AU when Neptune reached ~28 AU. If Neptune's semimajor axis changed by fraction of AU at this point, perhaps because Neptune was scattered off of another planet, the 2:1 population would have been released at ~44 AU, and would remain there to this day. We show that the orbital distribution of bodies produced in this model provides a good match to...
TORCH Computational Reference Kernels - A Testbed for Computer Science Research
Kaiser, Alex; Williams, Samuel Webb; Madduri, Kamesh; Ibrahim, Khaled; Bailey, David H.; Demmel, James W.; Strohmaier, Erich
2010-12-02T23:59:59.000Z
For decades, computer scientists have sought guidance on how to evolve architectures, languages, and programming models in order to improve application performance, efficiency, and productivity. Unfortunately, without overarching advice about future directions in these areas, individual guidance is inferred from the existing software/hardware ecosystem, and each discipline often conducts their research independently assuming all other technologies remain fixed. In today's rapidly evolving world of on-chip parallelism, isolated and iterative improvements to performance may miss superior solutions in the same way gradient descent optimization techniques may get stuck in local minima. To combat this, we present TORCH: A Testbed for Optimization ResearCH. These computational reference kernels define the core problems of interest in scientific computing without mandating a specific language, algorithm, programming model, or implementation. To compliment the kernel (problem) definitions, we provide a set of algorithmically-expressed verification tests that can be used to verify a hardware/software co-designed solution produces an acceptable answer. Finally, to provide some illumination as to how researchers have implemented solutions to these problems in the past, we provide a set of reference implementations in C and MATLAB.
An Implementation of Bayesian Adaptive Regression Splines (BARS)
Kass, Rob
extensive experience in fitting curves to neu- rophysiological data (Kass, Ventura, and Brown, 2005; Kass, Ventura, and Cai, 2003; Ventura et al., 2002). A typical data display is given in Figure 1. The raw data smoothers (Ventura et al., 2002), but we came across neurons like the one displayed in Figure 1 where kernel
Xiaowei Yang; Qing Shen; Hongquan Xu; Steven Shoptaw
2011-01-01T23:59:59.000Z
1997. 8. Faraway, JJ. Regression analysis for a functionalLocally Weighted Regression and Smoothing Scatterplots,Functional Regression Analysis using an F Test for
Penalized-regression-based multimarker genotype analysis of Genetic Analysis Workshop 17 data
Ayers, Kristin L; Mamasoula, Chrysovalanto; Cordell, Heather J
2011-01-01T23:59:59.000Z
via penalized logistic regression. Genet Epidemiol 2010, 34:The group LASSO for logistic regression. J R Stat Soc Ser Bvariants by penalized regression. Bioinformatics 2010, 26:
Regression Given input data (features), predict value of a
Giger, Christine
^ = (XX> ) 1 Xy etc. #12;! · Many relations are not linear · The complete graph Non-linear regression #12;! · Need to fit non-linear functions · example: polynomials Non-linear regression y = 6 10000 x5 82 10000 x) the dimension of the inputs · After applying z(x), we fit a plane in 3D-space Non-linear regression y = 0 + 1x
Doing data regression in the TI-85 Entering the data
Torres, Rodolfo
Doing data regression in the TI-85 Entering the data Press to get the statistics menu. Press (key F2). Select and and press .xStat yStat Enter the x-data and y-data. Doing the regression Press in the menu (key M1= 2nd F1).STAT Select and .xStat yStat Select the type of regression. For example, and get
Improving the Fisher Kernel for Large-Scale Image Classification
Kim, Tae-Kyun
classification the FK was shown to extend the popular bag-of-visual-words (BOV) by going beyond count statistics. However, in practice, this enriched representation has not yet shown its superiority over the BOV classification to date has been to describe images with bag-of-visual-words (BOV) histograms and to classify them
Terminologies Trivial extensions and representations
Iyama, Osamu
. For an artin algebra A over a commutative artinian ring C, D := HomC (-, E0 C (top C)), A = DTrA, -1 A = Tr of Algebra at University of Tsukuba The view from the old seminar at Univesity of Tsukuba Jun-ichi Miyachi November 14, 2013, Nagoya Jun-ichi Miyachi Researches on the Representation Theory of Algebra at Univers
Representation of Limited Rights Data and Restricted Computer...
Office of Energy Efficiency and Renewable Energy (EERE) Indexed Site
Representation of Limited Rights Data and Restricted Computer Software Representation of Limited Rights Data and Restricted Computer Software Representation of Limited Rights Data...
Using Enhanced Spherical Images for Object Representation
Smith, David A.
1979-05-01T23:59:59.000Z
The processes involved in vision, manipulation, and spatial reasoning depend greatly on the particular representation of three-dimensional objects used. A novel representation, based on concepts of differential geometry, ...
Collision kernels from velocity-selective optical pumping with magnetic depolarization
Bhamre, T.
We experimentally demonstrate how magnetic depolarization of velocity-selective optical pumping can be used to single out the collisional cusp kernel best describing spin- and velocity-relaxing collisions between potassium ...
Zhuang, Wei
2007-09-17T23:59:59.000Z
A three dimensional (3-D) capacitance extraction algorithm based on a kernel independent hierarchical method and geometric moments is described. Several techniques are incorporated, which leads to a better overall performance for arbitrary...
A Gaussian-Like Immersed Boundary Kernel with Improved Translational Invariance and Smoothness
Bao, Yuan-Xun; Peskin, Charles S
2015-01-01T23:59:59.000Z
The immersed boundary (IB) method is a general mathematical framework for studying problems involving fluid-structure interactions in which an elastic structure is immersed in a viscous incompressible fluid. In the IB formulation, the fluid described by Eulerian variables is coupled with the immersed structure described by Lagrangian variables via the use of the Dirac delta function. From a numerical standpoint, the Lagrangian force spreading and the Eulerian velocity interpolation are carried out by a regularized, compactly supported discrete delta function or kernel. Immersed-boundary kernels are derived from a certain set of postulates to achieve approximate grid translational invariance, interpolation accuracy and computational efficiency. In this note, we present a new 6-point immersed-boundary kernel that is $\\mathcal{C}^3$ and features a substantially improved translational invariance compared to other common IB kernels.
CPM: A Graph Pattern Matching Kernel with Diffusion for Accurate Graph Classification
Kansas, University of
(the Molecular Library Initiative project) that aims to determine and publicize the biological activ chemical structure data sets and have compared our methods to all major graph kernel functions that we know
Protein fold recognition by alignment of amino acid residues using kernelized dynamic time warping
Protein fold recognition by alignment of amino acid residues using kernelized dynamic time warping distances between proteins. This method shows significant improvement in protein fold recognition. Overall March 2014 Keywords: Protein sequence Fold recognition Alignment method Feature extraction
A Library for Locally Weighted Projection Regression --Supplementary Documentation --
problems: · The function to be learnt is non-linear. Otherwise having multiple local models is a waste of resources, and you should rather use ordinary linear regression, or partial least squares (PLS) for the caseA Library for Locally Weighted Projection Regression -- Supplementary Documentation -- Stefan
Model selection and estimation of a component in additive regression
Paris-Sud XI, Université de
on s and is based on non-asymptotic model selection methods. Given some linear spaces collection {Sm, m M}, we proposed and, among them, a widely used is the linear regression Z = µ + k i=1 iX(i) + (2) where µ;drawback of linear regression is its lack of flexibility for modeling more complex dependencies between Z
A DISTRIBUTION WITH GIVEN MARGINALS AND GIVEN REGRESSION CURVE
Cuadras, Carles M.
; and the possibility of using this construction to test nonlinear regression procedures and methods of estimation; mixture of distributions; nonlinear regression; extremal correla tions. AMS(1991) subject classification this construction to the nonlinear case. If ' is a monotone nonlinear function, satisfying some restrictions (e
FORECASTING WATER DEMAND USING CLUSTER AND REGRESSION ANALYSIS
Keller, Arturo A.
resources resulting in water stress. Effective water management a solution Supply side management Demand side management #12;Developing a regression equation based on cluster analysis for forecasting waterFORECASTING WATER DEMAND USING CLUSTER AND REGRESSION ANALYSIS by Bruce Bishop Professor of Civil
MULTIVARIATE REGRESSION S-ESTIMATORS FOR ROBUST ESTIMATION AND INFERENCE
Van Aelst, Stefan
1 MULTIVARIATE REGRESSION S-ESTIMATORS FOR ROBUST ESTIMATION AND INFERENCE Stefan Van Aelst-estimators for multivariate regression. We study the robustness of the estimators in terms of their breakdown point and in and multivariate location and scatter. Furthermore we develop a fast and robust bootstrap method
Inference for Covariate Adjusted Regression via Varying Coefficient Models 1
Müller, Hans-Georg
¸ENT¨URK AND HANS-GEORG M¨ULLER University of California, Davis We consider covariate adjusted regression (CAR consider a variant of (1), where one observes contaminated versions of predictors and response. Contamination of the variables in the regression model occurs through a multiplicative factor that is determined
Asymptotic Equivalence and Adaptive Estimation for Robust Nonparametric Regression
Zhou, Harrison Huibin
Asymptotic Equivalence and Adaptive Estimation for Robust Nonparametric Regression T. Tony Cai1 and Harrison H. Zhou2 University of Pennsylvania and Yale University Abstract Asymptotic equivalence theory. In this paper we develop asymptotic equivalence results for robust nonparametric regression with unbounded loss
Feature selection in high dimensional regression problems for genomics
Paris-Sud XI, Université de
Feature selection in high dimensional regression problems for genomics Julie Hamon1,2,3 , Clarisse, France julien.jacques@lifl.fr Abstract. In the context of genomic selection in animal breeding and "closed to real" datasets. Keywords: Feature selection, combinatorial optimization, regression, genomic. 1
Atomic Representations of Rank 2 Graph Algebras
Davidson, Ken
Atomic Representations of Rank 2 Graph Algebras Kenneth R. Davidson a , Stephen C. Power b , Dilian University, Lancaster LA1 4YF, U.K. Abstract We provide a detailed analysis of atomic -representations- posed into a direct sum or direct integral of irreducible atomic representations. The building blocks
IAI : Knowledge Representation John A. Bullinaria, 2005
Bullinaria, John
is a Knowledge Representation? 3. Requirements of a Knowledge Representation 4. Practical Aspects of Good5-5 Practical Aspects of Good Representations In practice, the theoretical requirements for good Dictionary provides as good a definition as any: knowledge, nolij, n. assured belief; that which is known
Context Representation for Web Search Results
Baeza-Yates, Ricardo
Context Representation for Web Search Results Jesús Vegas Department of Computer Science U. Valladolid Context Representation for Web Search Results 2 Outline Intro Web search results in the web site and Future work #12;Context Representation for Web Search Results 3 Introduction Searching the web is one
The effect of stress cracked and broken corn kernels on alkaline processing losses
Jackson, David Scott
1986-01-01T23:59:59.000Z
(~. 70, P&. 06). There were significant differences in COD and DML (KRATIO= 100) between highly damaged corn and the less damaged counterpart of the same hybrid. Stress cracked corn, however, only slightly increased COD and DML. The ease of pericarp... Sigruficance of Com and Cooking Parameters . . . LIST OF FIGURES Page Stress Crack, Pericarp, and Broken Kernel Damage of Corn . . Flow Chart of Procedutes and Differences Between Cook Methods I and H 21 24 Correlation between Thousand Kernel Weight...
Quantum Control and Representation Theory
A. Ibort; J. M. Pérez-Pardo
2012-03-11T23:59:59.000Z
A new notion of controllability for quantum systems that takes advantage of the linear superposition of quantum states is introduced. We call such notion von Neumann controllabilty and it is shown that it is strictly weaker than the usual notion of pure state and operator controlability. We provide a simple and effective characterization of it by using tools from the theory of unitary representations of Lie groups. In this sense we are able to approach the problem of control of quantum states from a new perspective, that of the theory of unitary representations of Lie groups. A few examples of physical interest and the particular instances of compact and nilpotent dynamical Lie groups are discussed.
PRODUCT REPRESENTATION IN LIGHTWEIGHT FORMATS FOR PRODUCT LIFECYCLE MANAGEMENT (PLM)
Rzepa, Henry S.
PRODUCT REPRESENTATION IN LIGHTWEIGHT FORMATS FOR PRODUCT LIFECYCLE MANAGEMENT (PLM) Lian Ding representation, markup method, Representation Information Registry/Repository, PLM, CAD model 1. INTRODUCTION. This paper gives a survey of the lightweight representations employed in product lifecycle management (PLM
Temporal Representation in Semantic Graphs
Levandoski, J J; Abdulla, G M
2007-08-07T23:59:59.000Z
A wide range of knowledge discovery and analysis applications, ranging from business to biological, make use of semantic graphs when modeling relationships and concepts. Most of the semantic graphs used in these applications are assumed to be static pieces of information, meaning temporal evolution of concepts and relationships are not taken into account. Guided by the need for more advanced semantic graph queries involving temporal concepts, this paper surveys the existing work involving temporal representations in semantic graphs.
Infinite-Dimensional Representations of 2-Groups
John C. Baez; Aristide Baratin; Laurent Freidel; Derek K. Wise
2011-02-09T23:59:59.000Z
A "2-group" is a category equipped with a multiplication satisfying laws like those of a group. Just as groups have representations on vector spaces, 2-groups have representations on "2-vector spaces", which are categories analogous to vector spaces. Unfortunately, Lie 2-groups typically have few representations on the finite-dimensional 2-vector spaces introduced by Kapranov and Voevodsky. For this reason, Crane, Sheppeard and Yetter introduced certain infinite-dimensional 2-vector spaces called "measurable categories" (since they are closely related to measurable fields of Hilbert spaces), and used these to study infinite-dimensional representations of certain Lie 2-groups. Here we continue this work. We begin with a detailed study of measurable categories. Then we give a geometrical description of the measurable representations, intertwiners and 2-intertwiners for any skeletal measurable 2-group. We study tensor products and direct sums for representations, and various concepts of subrepresentation. We describe direct sums of intertwiners, and sub-intertwiners - features not seen in ordinary group representation theory. We study irreducible and indecomposable representations and intertwiners. We also study "irretractable" representations - another feature not seen in ordinary group representation theory. Finally, we argue that measurable categories equipped with some extra structure deserve to be considered "separable 2-Hilbert spaces", and compare this idea to a tentative definition of 2-Hilbert spaces as representation categories of commutative von Neumann algebras.
SPAN-4. A Point-Kernel Shield Evaluation Code
Wallace, O.J. [Bettis Atomic Power Lab., West Mifflin, PA, (United States)
1992-03-16T23:59:59.000Z
SPAN4 calculates the fast neutron dose rate, thermal neutron flux, gamma-ray flux, dose rate, and energy-absorption rate in rectangular, cylindrical, and spherical geometries by integrating appropriate exponential kernels over a source distribution. The shield configuration is flexible, a first-level shield mesh, using any one of the three geometries, is specified. Regions of this same geometry or of other geometries, having their own (finer) meshes, may then be embedded between the first-level mesh lines, defining second-level shield meshes. This process is telescopic, third-level shield meshes may be embedded between second-level meshlines in turn. All meshes may have variable spacing. Sources and detectors may be located arbitrarily with respect to any shield mesh. The source is defined by the function: s=s0+s1(a)*s2(b)*s3(c)+s4(a,b)*s3(c)+s5(a,c)*s2(b)+s6(b,c) *s1(a)+s7(a,b,c), where a, b, and c represent coordinates. If any factor is missing, the corresponding terms are zero.
A multi-regression analysis of airline indirect operating costs
Taneja, Nawal K.
1968-01-01T23:59:59.000Z
A multiple regression analysis of domestic and local airline indirect costs was carried out to formulate cost estimating equations for airline indirect costs. Data from CAB and FAA sources covering the years 1962-66 was ...
Topics on Regularization of Parameters in Multivariate Linear Regression
Chen, Lianfu
2012-02-14T23:59:59.000Z
My dissertation mainly focuses on the regularization of parameters in the multivariate linear regression under different assumptions on the distribution of the errors. It consists of two topics where we develop iterative procedures to construct...
Regression analysis of oncology drug licensing deal values
Hawkins, Paul Allen
2006-01-01T23:59:59.000Z
This work is an attempt to explain wide variations in drug licensing deal value by using regression modeling to describe and predict the relationship between oncology drug deal characteristics and their licensing deal ...
Robust regression on noisy data for fusion scaling laws
Verdoolaege, Geert, E-mail: geert.verdoolaege@ugent.be [Department of Applied Physics, Ghent University, B-9000 Ghent (Belgium); Laboratoire de Physique des Plasmas de l'ERM - Laboratorium voor Plasmafysica van de KMS (LPP-ERM/KMS), Ecole Royale Militaire - Koninklijke Militaire School, B-1000 Brussels (Belgium)
2014-11-15T23:59:59.000Z
We introduce the method of geodesic least squares (GLS) regression for estimating fusion scaling laws. Based on straightforward principles, the method is easily implemented, yet it clearly outperforms established regression techniques, particularly in cases of significant uncertainty on both the response and predictor variables. We apply GLS for estimating the scaling of the L-H power threshold, resulting in estimates for ITER that are somewhat higher than predicted earlier.
Median regression and the missing information principle Ian W. McKeague
McKeague, Ian
Median regression and the missing information principle Ian W. McKeague Department of Statistics regression analysis has robustness properties which make it an attractive alternative to regression based of the median regression model, leading to a new estimator for the regression parameters. Our approach adapts
Regression Benchmarking with Simple Middleware Benchmarks Lubomr Bulej1,2
Regression Benchmarking with Simple Middleware Benchmarks Lubomír Bulej1,2 , Tomás Kalibera1 , Petr.tuma}@mff.cuni.cz Abstract The paper introduces the concept of regression benchmarking as a variant of regression testing fo- cused at detecting performance regressions. Applying the regression benchmarking in the area of middle
Scattering in flatland: Efficient representations via wave atoms
Peraire, Jaime
2009 Abstract This paper presents a numerical compression strategy for the boundary integral equation of acoustic scattering in two dimensions. These equations have oscillatory kernels that we represent of the kernel of the double-layer potential for this scatterer, sampled as a 1024x1024 matrix. Right: a zoomed
KNOWMESH --Meshless geometry with knowledge representation
Franklin, W. Randolph
(aka path, motion) planning · Drainage analysis Commercial product's inconsistent layer representation setup Closeup of tank Dense grid of scanner points · model with homogeneous lifts
ON AUTOMORPHY OF CERTAIN GALOIS REPRESENTATIONS OF ...
2015-02-09T23:59:59.000Z
22. 1991 Mathematics Subject Classification. Primary 14F30,14L05. Key words and phrases. Galois representations, automorphy. This materials is based upon ...
Knowledge and Skill Representations for Robotized Production
Malec, Jacek
: Model-based control, Knowledge representation, System architectures, Autonomous control, Industrial robots 1. INTRODUCTION Model-based systems in control are a means to utilize effi- ciently human
dos Santos, Pedro G.
2012-12-31T23:59:59.000Z
This dissertation provides insights on what influences women's descriptive representation in state legislatures in Brazil. The study of female representation in Brazil provides for a good case study as the country uses a gender quota system...
Lindemer, Terrence [Harbach Engineering and Solutions] [Harbach Engineering and Solutions; Voit, Stewart L [ORNL] [ORNL; Silva, Chinthaka M [ORNL] [ORNL; Besmann, Theodore M [ORNL] [ORNL; Hunt, Rodney Dale [ORNL] [ORNL
2014-01-01T23:59:59.000Z
The U.S. Department of Energy is considering a new nuclear fuel that would be less susceptible to ruptures during a loss-of-coolant accident. The fuel would consist of tristructural isotropic coated particles with large, dense uranium nitride (UN) kernels. This effort explores many factors involved in using gel-derived uranium oxide-carbon microspheres to make large UN kernels. Analysis of recent studies with sufficient experimental details is provided. Extensive thermodynamic calculations are used to predict carbon monoxide and other pressures for several different reactions that may be involved in conversion of uranium oxides and carbides to UN. Experimentally, the method for making the gel-derived microspheres is described. These were used in a microbalance with an attached mass spectrometer to determine details of carbothermic conversion in argon, nitrogen, or vacuum. A quantitative model is derived from experiments for vacuum conversion to an uranium oxide-carbide kernel.
Arhonditsis, George B.
Multiple regression models: A methodology for evaluating trihalomethane concentrations in drinking of these substances on human health. A multiple regression model was developed to estimate THM concentrations Science Ltd. All rights reserved. Keywords: Multiple regression model; Trihalomethanes; Drinking water
Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression
Chen, Wan-Ping
2015-01-01T23:59:59.000Z
in Multi-response Linear Regression Problem Set-up and theList of Figures The linear regression model Y ? X?, with p nin high-dimensional multivariate regression. ” The Annals of
Debray, Saumya
-purpose operating systems on embedded platforms. The problem is complicated by the fact that kernel code tends imple- mentation of our ideas on an Intel x86 platform, applied to a Linux kernel that has been will typically not have a mouse interface); at the software end, they usually support a fixed set of applications
Heat kernels on 2d Liouville quantum gravity: a numerical study
Grigory Bonik; Joe P. Chen; Alexander Teplyaev
2014-11-06T23:59:59.000Z
We numerically compute the heat kernel on a square lattice torus equipped with the measure corresponding to Liouville quantum gravity (LQG). From the on-diagonal heat kernel we verify that the spectral dimension of LQG is 2. Furthermore, when diffusion is started from a high point of the underlying Gaussian free field, our numerics indicates superdiffusive space-time scaling with respect to the Euclidean metric in the small space-to-time regime. The implications of this result require further investigation, but seem to coincide with the notion that the Euclidean metric is not the right geodesic for characterizing the geometry of LQG.
Topics in Representation Theory: The Heisenberg Algebra
Woit, Peter
Topics in Representation Theory: The Heisenberg Algebra We'll now turn to a topic which is a precise analog of the previous discussion of the Clifford algebra and spinor representations. By replacing a new algebra, the Heisenberg algebra. The group of automor- phism of this algebra is now a symplectic
Representations of Petri net interactions Pawel Sobocinski
Sobocinski, Pawel
Representations of Petri net interactions Pawel SobociÂ´nski ECS, University of Southampton, UK Abstract. We introduce a novel compositional algebra of Petri nets, as well as a stateful extension In part owing to their intuitive graphical representation, Petri nets [28] are of- ten used both
N-representability is QMA-complete
Y. -K. Liu; M. Christandl; F. Verstraete
2006-09-17T23:59:59.000Z
We study the computational complexity of the N-representability problem in quantum chemistry. We show that this problem is QMA-complete, which is the quantum generalization of NP-complete. Our proof uses a simple mapping from spin systems to fermionic systems, as well as a convex optimization technique that reduces the problem of finding ground states to N-representability.
Knowledge representation in process engineering Ulrike Sattler
Baader, Franz
Knowledge representation in process engineering Ulrike Sattler RWTH Aachen, uli, the tasks we are concerned with in process engineering are described as well as how knowledge representation@cantor.informatik.rwthaachen.de Abstract Process engineering is surely no pure con figuration application, but modeling the structure
Rational Univariate Representation 1 Stickelberger's Theorem
Verschelde, Jan
(RUR) 2 The Elbow Manipulator a spatial robot arm with three links 3 Application of the Newton a rational univariate representation (RUR) 2 The Elbow Manipulator a spatial robot arm with three links 3 a rational univariate representation (RUR) 2 The Elbow Manipulator a spatial robot arm with three links 3
Shell Element Verification & Regression Problems for DYNA3D
Zywicz, E
2008-02-01T23:59:59.000Z
A series of quasi-static regression/verification problems were developed for the triangular and quadrilateral shell element formulations contained in Lawrence Livermore National Laboratory's explicit finite element program DYNA3D. Each regression problem imposes both displacement- and force-type boundary conditions to probe the five independent nodal degrees of freedom employed in the targeted formulation. When applicable, the finite element results are compared with small-strain linear-elastic closed-form reference solutions to verify select aspects of the formulations implementation. Although all problems in the suite depict the same geometry, material behavior, and loading conditions, each problem represents a unique combination of shell formulation, stabilization method, and integration rule. Collectively, the thirty-six new regression problems in the test suite cover nine different shell formulations, three hourglass stabilization methods, and three families of through-thickness integration rules.
UNCORRECTED 2 Kernel-based distance metric learning for content-based image retrieval
Yeung, Dit-Yan
UNCORRECTED PROOF 1 2 Kernel-based distance metric learning for content-based image retrieval 3 and Technology, Clear Water Bay, Kowloon, Hong Kong 5 Received 24 August 2005; received in revised form 16 images, the performance of a content-based image retrieval (CBIR) system 9 depends critically
SPEK: A Storage Performance Evaluation Kernel Module for Block Level Storage Systems
He, Xubin "Ben"
SPEK: A Storage Performance Evaluation Kernel Module for Block Level Storage Systems Ming Zhang Cookeville, TN 38505, USA hexb@tntech.edu Abstract In this paper we introduce SPEK (Storage Performance storage systems at block level. It can be used for both DAS (Direct Attached Storage) and block level
Structure of an Aspergillus flavus population from maize kernels in northern Italy Antonio Mauro a
Cotty, Peter J.
; Piva et al., 2006). Aflatoxins are secondary metabolites produced by several mem- bers of AspergillusStructure of an Aspergillus flavus population from maize kernels in northern Italy Antonio Mauro, a Institute of Entomology and Plant Pathology, Universitŕ Cattolica del Sacro Cuore, Via Emilia Parmense 84
Yang, Junfeng
page frame sharing can be leveraged for the complete circumven- tion of software and hardware kernel, and kGuard. We also discuss techniques for constructing reliable ret2dir exploits against x86, x86-level software has become much harder, as recent versions of popular OSes come with nu- merous protections
Shi, Tao
Abstract Polar Cloud Detection using Satellite Data with Analysis and Application of Kernel Professor Bin Yu, Chair Clouds play a major role in Earth's climate and cloud detection is a crucial step climate model studies. Cloud detection is particularly difficult in the snow- and ice-covered polar
Viability Kernel for Ecosystem Management Models Eladio Oca~na Anaya
Paris-Sud XI, Université de
Viability Kernel for Ecosystem Management Models Eladio Oca~na Anaya Michel De Lara Ricardo task in general. We study the viability of nonlinear generic ecosystem models under preservation in the Peruvian upwelling ecosystem. Key words: control theory; state constraints; viability; predator
Aguiar, Pedro M. Q.
COMBINING FREE ENERGY SCORE SPACES WITH INFORMATION THEORETIC KERNELS: APPLICATION TO SCENE recent and top performing tools in each of the steps: (i) the free energy score space; (ii) non embeddings can be found in [1, 2, 13]. A very recent approach, termed free energy score space (FESS) [13, 14
Predictable Interrupt Management for Real Time Kernels over conventional PC Hardware1
Mejia-Alvarez, Pedro
which this integrated model improves the traditional model. The design of a flexible and portable kernel-CONACyT 42151-Y, and CONACYT 42449-Y Mexico. Abstract In this paper we analyze the traditional model on real-time systems. As a result of this analysis, we propose a model that integrates interrupts
EXTENSIONS OF LINKING SYSTEMS WITH p-GROUP KERNEL BOB OLIVER AND JOANA VENTURA
Ventura, Joana
EXTENSIONS OF LINKING SYSTEMS WITH p-GROUP KERNEL BOB OLIVER AND JOANA VENTURA Abstract. We study. Oliver is partially supported by UMR 7539 of the CNRS. J. Ventura is partially supported by FCT. #12;2 BOB OLIVER AND JOANA VENTURA enother prolem is tht in generlD when ev is linking system nd A g
EXTENSIONS OF LINKING SYSTEMS WITH p-GROUP KERNEL BOB OLIVER AND JOANA VENTURA
Oliver, Bob
AND JOANA VENTURA Abstract. We study extensions of p-local finite groups where the kernel i* *s of the CNRS. J. Ventura is partially supported by FCT/POCTI/FEDER and grant PDCT/MAT/58497* */2004. Both;2 BOB OLIVER AND JOANA VENTURA Another problem is that in general, when eLis a linking system and A C e
AN EFFICIENT INTEGRAL TRANSFORM TECHNIQUE OF A SINGULAR WIRE ANTENNA KERNEL
Park, Seong-Ook
AN EFFICIENT INTEGRAL TRANSFORM TECHNIQUE OF A SINGULAR WIRE ANTENNA KERNEL S.-O. Park Department-348, South of Korea Abstract-This paper presents an efficient integral transform tech- nique for evaluating transforma- tions, the original double integral 1/-Rs with a singular kernal can be represented as a finite
Integral operators with the generalized sine-kernel on the real axis
N. A. Slavnov
2010-05-27T23:59:59.000Z
The asymptotic properties of integral operators with the generalized sine kernel acting on the real axis are studied. The formulas for the resolvent and the Fredholm determinant are obtained in the large x limit. Some applications of the results obtained to the theory of integrable models are considered.
Optimizing Kernel Block Memory Operations Michael Calhoun, Scott Rixner, Alan L. Cox
Optimizing Kernel Block Memory Operations Michael Calhoun, Scott Rixner, Alan L. Cox Rice University Houston, TX 77005 {calhoun,rixner,alc}@rice.edu Abstract-- This paper investigates the performance of block memory operations in the operating system, including memory copies, page zeroing, interprocess
Global Heat Kernel Estimate for Relativistic Stable Processes in Exterior Open Sets
Chen, Zhen-Qing
,1 exterior open sets as well as for half-space-like open sets. The ideas of [8] have been adaptedGlobal Heat Kernel Estimate for Relativistic Stable Processes in Exterior Open Sets Zhen-Qing Chen for the transition densities of relativistic -stable processes with mass m (0, 1] in C1,1 exterior open sets
IEEE SIGNAL PROCESSING LETTER, VOL. , NO. , 2008 1 Image interpolation by blending kernels
interpolation kernel function cannot yield high quality high resolution(HR for short) images in practice, since before being used to interpolate the low resolution(LR for short) image. The main problem is that if we of values at control points to a spline function. This method is also used in practice currently
Using a Secure Java Micro-kernel on Embedded Devices for the Reliable Execution of
Binder, Walter
Using a Secure Java Micro-kernel on Embedded Devices for the Reliable Execution of Dynamically Uploaded Applications Walter Binder and Bal´azs Lichtl CoCo Software Engineering GmbH Margaretenstr. 22 applications. Mobile code is used for application upload, as well as for remote configuration and maintenance
Gene Feature Extraction Using T-Test Statistics and Kernel Partial Least Squares
Kwok, James Tin-Yau
Gene Feature Extraction Using T-Test Statistics and Kernel Partial Least Squares Shutao Li1 , Chen Clear Water Bay, Hong Kong shutao li@yahoo.com.cn, lc337199@sina.com, jamesk@cs.ust.hk Abstract. In this paper, we propose a gene extraction method by us- ing two standard feature extraction methods, namely
A Kernel-Based Spatio-Temporal Dynamical Model for Nowcasting Weather Radar Reflectivities
A Kernel-Based Spatio-Temporal Dynamical Model for Nowcasting Weather Radar Reflectivities Ke Xu of the technique and its potential for nowcasting weather radar reflectivities. Key Words: Bayesian, dilation to nowcasting weather radar reflectivities into two general categories. The first is the use of simple
Vasconcelos, Nuno M.
Supplemental Material for ``Classifying Video with Kernel Dynamic Textures'' Antoni B. Chan and Nuno Vasconcelos SVCLTR 2007/03 April 2007 #12; #12; Supplemental Material for ``Classifying Video This is the supplemental material for ``Classifying Video with Dynamic Textures'' [1]. It contains information about
AIRCRAFT PARAMETRIC STRUCTURAL LOAD MONITORING USING GAUSSIAN PROCESS REGRESSION
Boyer, Edmond
cases. KEYWORDS : Structural Health and Usage Monitoring, Gaussian Process Regression, Fatigue, 1 in the remaining useful life. If the error is too 7th European Workshop on Structural Health Monitoring July 8 manuscript, published in "EWSHM - 7th European Workshop on Structural Health Monitoring (2014)" #12
Storage Device Performance Prediction with Selective Bagging Classification and Regression
Paris-Sud XI, Université de
Storage Device Performance Prediction with Selective Bagging Classification and Regression Tree Lei}@eng.wayne.edu, cheneh@ustc.edu.cn Abstract. Storage device performance prediction is a key element of self-managed storage systems and application planning tasks, such as data assignment and configuration. Based
1Machine Learning, to appear. Least-Squares Independence Regression
Sugiyama, Masashi
1Machine Learning, to appear. Least-Squares Independence Regression for Non-Linear Causal Inference of Technology, Japan. sesejun@cs.titech.ac.jp Abstract The discovery of non-linear causal relationship under Causal inference, Non-Linear, Non-Gaussian, Squared-loss mutual information, Least-Squares Independence
Efficient Locally Weighted Polynomial Regression Predictions Andrew W. Moore
Schneider, Jeff
polynomial regression (LWPR) is a popular instancebased al gorithm for learning continuous nonlinear gorithm for learning continuous nonlinear mappings from realvalued input vectors to realvalued output vectors. It is particularly appropriate for learning com plex highly nonlinear functions of up to about
Climate Multi-model Regression Using Spatial Smoothing Karthik Subbian
Banerjee, Arindam
Climate Multi-model Regression Using Spatial Smoothing Karthik Subbian Arindam Banerjee Abstract There are several Global Climate Models (GCMs) reported by var- ious countries to the Intergovernmental Panel on Climate Change (IPCC). Due to the varied nature of the GCM assumptions, the fu- ture projections
Breakdown points of Cauchy regression-scale estimators Ivan Mizera
Mizera, Ivan
@stat.ualberta.ca. This work was supported in part by the National Sciences and Engineering Research Council of Canada. 2 of Mathematical and Statistical Sciences, University of Alberta, Edmonton, Alberta, T6G 2G1, Canada. Email: mizeraBreakdown points of Cauchy regression-scale estimators Ivan Mizera University of Alberta1
Polynomial regression with derivative information in nuclear reactor uncertainty quantification*
Anitescu, Mihai
parameters on the performance of a model of sodium-cooled fast reactor. The experiments show-cooled fast reactor. We construct a surrogate model as a goal-oriented projection onto an incomplete space1 Polynomial regression with derivative information in nuclear reactor uncertainty quantification
Worldwide Oil Production Michaelis-Menten Kinetics Correlation and Regression
Watkins, Joseph C.
Michaelis-Menten Kinetics Worldwide Oil Production Example. The modern history of petroleum began in the 19Worldwide Oil Production Michaelis-Menten Kinetics Topic 4 Correlation and Regression Transformed Variables 1 / 13 #12;Worldwide Oil Production Michaelis-Menten Kinetics Outline Worldwide Oil Production
Efficiency transfer for regression models with responses missing at random
Mueller, Uschi
Efficiency transfer for regression models with responses missing at random Ursula U. M that characteristics of the con- ditional distribution of Y given X can be estimated efficiently using complete case analysis. One can simply omit incomplete cases and work with an appro- priate efficient estimator without
Evaluation of Regressive Methods for Automated Generation of Test Trajectories
Cukic, Bojan
. In simple terms,a test trajectoiy is defined as a series of data points, with each (possiblymultidimensional of test generation and the percentage of "acceptable" trajectories, measured by the domain specificEvaluation of Regressive Methods for Automated Generation of Test Trajectories Brian J. Taylor
Bootstrap Tests for Overidentification in Linear Regression Models
Spino, Claude
Bootstrap Tests for Overidentification in Linear Regression Models Russell Davidson Department it impossible to perform reliable inference near the point at which the limit is ill-defined. Several bootstrap are not too weak. We also study the power properties of the bootstrap tests. JEL codes: C10, C12, C15, C30
DYNA3D/ParaDyn Regression Test Suite Inventory
Lin, J I
2011-01-25T23:59:59.000Z
The following table constitutes an initial assessment of feature coverage across the regression test suite used for DYNA3D and ParaDyn. It documents the regression test suite at the time of production release 10.1 in September 2010. The columns of the table represent groupings of functionalities, e.g., material models. Each problem in the test suite is represented by a row in the table. All features exercised by the problem are denoted by a check mark in the corresponding column. The definition of ''feature'' has not been subdivided to its smallest unit of user input, e.g., algorithmic parameters specific to a particular type of contact surface. This represents a judgment to provide code developers and users a reasonable impression of feature coverage without expanding the width of the table by several multiples. All regression testing is run in parallel, typically with eight processors. Many are strictly regression tests acting as a check that the codes continue to produce adequately repeatable results as development unfolds, compilers change and platforms are replaced. A subset of the tests represents true verification problems that have been checked against analytical or other benchmark solutions. Users are welcomed to submit documented problems for inclusion in the test suite, especially if they are heavily exercising, and dependent upon, features that are currently underrepresented.
Non-smooth brownian martingales and stochastic integral representations
Wroblewski, David M.
2007-01-01T23:59:59.000Z
and M. Yor. On stochastic integral representations ofmotion by stochas- tic integrals. Ann. Math. Statist. , 41:Martingales and Stochastic Integral Representations A
West, Mike
classification, validation, prognosis Binary regression models · Linear regression model based on regression Standard statistical models transform from real-value to (0, 1) using a specified non-linear functionStatistics & Gene Expression Data Analysis Note 8: Binary Regression Outcomes and classification
arXiv:submit/0910499[stat.ML]11Feb2014 Online Nonparametric Regression
Rakhlin, Alexander "Sasha"
learning with squared loss and online nonparametric regression are the same. In addition to a non experts and for online linear regression. 1 Introduction Within the online regression framework, data (x1, starting with the paper of Foster [8], has been almost exclusively on finite-dimensional linear regression
A note on wavelet estimation of the derivatives of a regression function in a
Paris-Sud XI, Université de
A note on wavelet estimation of the derivatives of a regression function in a random design setting of the derivatives of a regression function in the nonparametric regression model with random design. New wavelet. Keywords and phrases: Nonparametric regression, Derivatives function estimation, Wavelets, Besov balls
Doing data regression in the TI82 or TI83 Entering the data
Torres, Rodolfo
Doing data regression in the TI82 or TI83 Entering the data ffl Press STAT to get the statistics the regression ffl Press STAT to get the statistics menu. ffl Select CALC and select the type of regression). ffl Press ENTER and get the answer. Plotting the data and the regression curve ffl Clear the Y= of any
Path Thresholding: Asymptotically Tuning-Free High-Dimensional Sparse Regression
Path Thresholding: Asymptotically Tuning-Free High-Dimensional Sparse Regression Divyanshu Vats of selecting tuning parame- ters for high-dimensional sparse regression. We propose a simple and computation-dependent sparse regression algorithm into an asymptotically tuning-free sparse regression algorithm. More specifi
Regression Test Selection for AspectJ Software Ohio State University
Xu, Guoqing "Harry"
Regression Test Selection for AspectJ Software Guoqing Xu Ohio State University Atanas Rountev Ohio, the new program needs to be regression tested to validate these changes. To reduce the cost of regression test- ing, a regression-test-selection technique can be used to se- lect only a necessary subset
Use of Regression Equations 1 Running head: Equations from summary data
Crawford, John R.
Use of Regression Equations 1 Running head: Equations from summary data Neuropsychology, in press the final version published in the APA journal. It is not the copy of record Using regression equations.crawford@abdn.ac.uk #12;Use of Regression Equations 2 Abstract Regression equations have many useful roles
Hu, Yaozhong; Nualart, David
2009-11-09T23:59:59.000Z
argument can be made easily by approximating the Dirac delta function by the heat kernel p"(x) = 1p 2pi" e ?x2/2" as " tends to zero. That is, Gt(h) is the limit in L2(?) as " tends to zero of G"t (h) =?2 ? t 0 ? v 0 #0; p"(Bv ? Bu + h) + p"(Bv ? Bu ? h)? 2...p"(Bv ? Bu) #1; dudv. (2.6) Applying Clark-Ocone formula we can derive the following stochastic integral representation for Gt(h). Proposition 2. The random variable Gt(h) defined in (1.3) can be expressed as Gt(h) = E(Gt(h)) + ? t 0 ut...
Representation, Organization, Classification, and Meaning-Making
Toronto, University of
1 Representation, Organization, Classification, and Meaning-Making Description Fundamental epistemological and ontological issues in the use of knowledge and information in human activities. Analysis, department store, grocery store, children's library, a menu, a store catalogue) and analyze that organization
Mental Representations Formed From Educational Website Formats
Elizabeth T. Cady; Kimberly R. Raddatz; Tuan Q. Tran; Bernardo de la Garza; Peter D. Elgin
2006-10-01T23:59:59.000Z
The increasing popularity of web-based distance education places high demand on distance educators to format web pages to facilitate learning. However, limited guidelines exist regarding appropriate writing styles for web-based distance education. This study investigated the effect of four different writing styles on reader’s mental representation of hypertext. Participants studied hypertext written in one of four web-writing styles (e.g., concise, scannable, objective, and combined) and were then administered a cued association task intended to measure their mental representations of the hypertext. It is hypothesized that the scannable and combined styles will bias readers to scan rather than elaborately read, which may result in less dense mental representations (as identified through Pathfinder analysis) relative to the objective and concise writing styles. Further, the use of more descriptors in the objective writing style will lead to better integration of ideas and more dense mental representations than the concise writing style.
Lyapunov Function Synthesis using Handelman Representations.
Sankaranarayanan, Sriram
Lyapunov Function Synthesis using Handelman Representations. Sriram Sankaranarayanan Xin Chen investigate linear programming relaxations to synthesize Lyapunov functions that es- tablish the stability approach searches for a Lyapunov function, given a parametric form with unknown coefficients
Towards improving phenotype representation in OWL
Loebe, Frank; Stumpf, Frank; Hoehndorf, Robert; Herre, Heinrich
2012-09-21T23:59:59.000Z
PROCEEDINGS Open Access Towards improving phenotype representation in OWL Frank Loebe1*, Frank Stumpf1, Robert Hoehndorf2, Heinrich Herre3 From Ontologies in Biomedicine and Life Sciences (OBML 2011) Berlin, Germany. 6-7 October 2011...
Text representations in digital hypermedia library systems
Lokken, Sveinung Taraldsrud
1993-01-01T23:59:59.000Z
The advent of the digital library poses a great number of challenging research questions in the areas of hypermedia, computer-human interaction, information retrieval, and information science. Choosing a representation for text converted from...
Representations up to homotopy of Lie algebroids
Abad, Camilo Arias
2009-01-01T23:59:59.000Z
This is the first in a series of papers devoted to the study of the cohomology of classifying spaces. The aim of this paper is to introduce and study the notion of representation up to homotopy and to make sense of the adjoint representation of a Lie algebroid. Our construction is inspired by Quillen's notion of superconnection and fits into the general theory of structures up to homotopy. The advantage of considering such representations is that they are flexible and general enough to contain interesting examples which are the correct generalization of the corresponding notions for Lie algebras. They also allow one to identify seemingly ad-hoc constructions and cohomology theories as instances of the cohomology with coefficients in representations (up to homotopy). In particular, we show that the adjoint representation of a Lie algebroid makes sense as a representation up to homotopy and that, similar to the case of Lie algebras, the resulting cohomology controls the deformations of the Lie algebroid (i.e. i...
Comparing Single and Multiple Turbine Representations in a Wind Farm Simulation: Preprint
Muljadi, E.; Parsons, B.
2006-03-01T23:59:59.000Z
This paper compares single turbine representation versus multiple turbine representation in a wind farm simulation.
PROPERTIES OF A SOLAR FLARE KERNEL OBSERVED BY HINODE AND SDO
Young, P. R. [College of Science, George Mason University, 4400 University Drive, Fairfax, VA 22030 (United States)] [College of Science, George Mason University, 4400 University Drive, Fairfax, VA 22030 (United States); Doschek, G. A.; Warren, H. P. [Naval Research Laboratory, 4555 Overlook Avenue SW, Washington, DC 20375 (United States)] [Naval Research Laboratory, 4555 Overlook Avenue SW, Washington, DC 20375 (United States); Hara, H. [National Astronomical Observatory of Japan/NINS, 2-21-1 Osawa, Mitaka, Tokyo 181-8588 (Japan)] [National Astronomical Observatory of Japan/NINS, 2-21-1 Osawa, Mitaka, Tokyo 181-8588 (Japan)
2013-04-01T23:59:59.000Z
Flare kernels are compact features located in the solar chromosphere that are the sites of rapid heating and plasma upflow during the rise phase of flares. An example is presented from a M1.1 class flare in active region AR 11158 observed on 2011 February 16 07:44 UT for which the location of the upflow region seen by EUV Imaging Spectrometer (EIS) can be precisely aligned to high spatial resolution images obtained by the Atmospheric Imaging Assembly (AIA) and Helioseismic and Magnetic Imager (HMI) on board the Solar Dynamics Observatory (SDO). A string of bright flare kernels is found to be aligned with a ridge of strong magnetic field, and one kernel site is highlighted for which an upflow speed of Almost-Equal-To 400 km s{sup -1} is measured in lines formed at 10-30 MK. The line-of-sight magnetic field strength at this location is Almost-Equal-To 1000 G. Emission over a continuous range of temperatures down to the chromosphere is found, and the kernels have a similar morphology at all temperatures and are spatially coincident with sizes at the resolution limit of the AIA instrument ({approx}<400 km). For temperatures of 0.3-3.0 MK the EIS emission lines show multiple velocity components, with the dominant component becoming more blueshifted with temperature from a redshift of 35 km s{sup -1} at 0.3 MK to a blueshift of 60 km s{sup -1} at 3.0 MK. Emission lines from 1.5-3.0 MK show a weak redshifted component at around 60-70 km s{sup -1} implying multi-directional flows at the kernel site. Significant non-thermal broadening corresponding to velocities of Almost-Equal-To 120 km s{sup -1} is found at 10-30 MK, and the electron density in the kernel, measured at 2 MK, is 3.4 Multiplication-Sign 10{sup 10} cm{sup -3}. Finally, the Fe XXIV {lambda}192.03/{lambda}255.11 ratio suggests that the EIS calibration has changed since launch, with the long wavelength channel less sensitive than the short wavelength channel by around a factor two.
Complete Representations in Algebraic Logic Robin Hirsch & Ian Hodkinson \\Lambda
Hodkinson, Ian
Complete Representations in Algebraic Logic Robin Hirsch & Ian Hodkinson \\Lambda Abstract A boolean algebra is shown to be completely representable if and only if it is atomic, whereas it is shown that neither the class of completely representable relation algebras nor the class of completely representable
357 09/2010 Assessing Representations for Moving Object
Güting, Ralf Hartmut
INFORMATIK BERICHTE 357 09/2010 Assessing Representations for Moving Object Histories Christian Hagen #12;1 Assessing Representations for Moving Object Histories Christian D¨untgen, Thomas Behr of the histories of moving objects in databases: the Compact Representation, the Unit Representation and the Hybrid
CrossWalk: A Tool for Performance Profiling Across the UserKernel Boundary (ParCo Preprint)
Miller, Barton P.
significant amount of their time in the operating system. As a result, conventional userlevel profilers canWalk. By drilling down into the kernel, we were able to identify the ultimate cause of its performance problems
Poisson Regression Analysis of Illness and Injury Surveillance Data
Frome E.L., Watkins J.P., Ellis E.D.
2012-12-12T23:59:59.000Z
The Department of Energy (DOE) uses illness and injury surveillance to monitor morbidity and assess the overall health of the work force. Data collected from each participating site include health events and a roster file with demographic information. The source data files are maintained in a relational data base, and are used to obtain stratified tables of health event counts and person time at risk that serve as the starting point for Poisson regression analysis. The explanatory variables that define these tables are age, gender, occupational group, and time. Typical response variables of interest are the number of absences due to illness or injury, i.e., the response variable is a count. Poisson regression methods are used to describe the effect of the explanatory variables on the health event rates using a log-linear main effects model. Results of fitting the main effects model are summarized in a tabular and graphical form and interpretation of model parameters is provided. An analysis of deviance table is used to evaluate the importance of each of the explanatory variables on the event rate of interest and to determine if interaction terms should be considered in the analysis. Although Poisson regression methods are widely used in the analysis of count data, there are situations in which over-dispersion occurs. This could be due to lack-of-fit of the regression model, extra-Poisson variation, or both. A score test statistic and regression diagnostics are used to identify over-dispersion. A quasi-likelihood method of moments procedure is used to evaluate and adjust for extra-Poisson variation when necessary. Two examples are presented using respiratory disease absence rates at two DOE sites to illustrate the methods and interpretation of the results. In the first example the Poisson main effects model is adequate. In the second example the score test indicates considerable over-dispersion and a more detailed analysis attributes the over-dispersion to extra-Poisson variation. The R open source software environment for statistical computing and graphics is used for analysis. Additional details about R and the data that were used in this report are provided in an Appendix. Information on how to obtain R and utility functions that can be used to duplicate results in this report are provided.
Effect of Pre and Post-Harvest Treatments on Characteristics of ‘Pawnee’ Pecan Kernels
Mansur, Zainab J
2014-04-17T23:59:59.000Z
(Venkatachalam, 2004). Pecans are considered a healthy food because of the high monounsaturated fatty acid content (Villarreal-Lozoya et al., 2007) and high concentrations of phenolics, flavonoids, and proanthocyanidins, which are phytochemicals with strong...., 2004). Pecan kernels have a wide range of possible uses. They can be sold in shell or shelled, and used as a main ingredient for confectionery, dairy and bakery products. Other uses include: incorporation into snack bars in raw form, sweetening...
Unifying Geometrical Representations of Gauge Theory
Scott T Alsid; Mario A Serna
2014-10-28T23:59:59.000Z
We unify three approaches within the vast body of gauge-theory research that have independently developed distinct representations of a geometrical surface-like structure underlying the vector-potential. The three approaches that we unify are: those who use the compactified dimensions of Kaluza-Klein theory, those who use Grassmannian models (also called gauge theory embedding or $CP^{N-1}$ models) to represent gauge fields, and those who use a hidden spatial metric to replace the gauge fields. In this paper we identify a correspondence between the geometrical representations of the three schools.Each school was mostly independently developed, does not compete with other schools, and attempts to isolate the gauge-invariant geometrical surface-like structures that are responsible for the resulting physics. By providing a mapping between geometrical representations, we hope physicists can now isolate representation-dependent physics from gauge-invariant physical results and share results between each school. We provide visual examples of the geometrical relationships between each school for $U(1)$ electric and magnetic fields. We highlight a first new result: in all three representations a static electric field (electric field from a fixed ring of charge or a sphere of charge) has a hidden gauge-invariant time dependent surface that is underlying the vector potential.
Fission product release and survivability of UN-kernel LWR TRISO fuel
T. M. Besmann; M. K. Ferber; H.-T. Lin; B. P. Collin
2014-05-01T23:59:59.000Z
A thermomechanical assessment of the LWR application of TRISO fuel with UN kernels was performed. Fission product release under operational and transient temperature conditions was determined by extrapolation from fission product recoil calculations and limited data from irradiated UN pellets. Both fission recoil and diffusive release were considered and internal particle pressures computed for both 650 and 800 um diameter kernels as a function of buffer layer thickness. These pressures were used in conjunction with a finite element program to compute the radial and tangential stresses generated within a TRISO particle undergoing burnup. Creep and swelling of the inner and outer pyrolytic carbon layers were included in the analyses. A measure of reliability of the TRISO particle was obtained by computing the probability of survival of the SiC barrier layer and the maximum tensile stress generated in the pyrolytic carbon layers from internal pressure and thermomechanics of the layers. These reliability estimates were obtained as functions of the kernel diameter, buffer layer thickness, and pyrolytic carbon layer thickness. The value of the probability of survival at the end of irradiation was inversely proportional to the maximum pressure.
Phenomenological memory-kernel master equations and time-dependent Markovian processes
L. Mazzola; E. -M. Laine; H. -P. Breuer; S. Maniscalco; J. Piilo
2011-03-03T23:59:59.000Z
Do phenomenological master equations with memory kernel always describe a non-Markovian quantum dynamics characterized by reverse flow of information? Is the integration over the past states of the system an unmistakable signature of non-Markovianity? We show by a counterexample that this is not always the case. We consider two commonly used phenomenological integro-differential master equations describing the dynamics of a spin 1/2 in a thermal bath. By using a recently introduced measure to quantify non-Markovianity [H.-P. Breuer, E.-M. Laine, and J. Piilo, Phys. Rev. Lett. 103, 210401 (2009)] we demonstrate that as far as the equations retain their physical sense, the key feature of non-Markovian behavior does not appear in the considered memory kernel master equations. Namely, there is no reverse flow of information from the environment to the open system. Therefore, the assumption that the integration over a memory kernel always leads to a non-Markovian dynamics turns out to be vulnerable to phenomenological approximations. Instead, the considered phenomenological equations are able to describe time-dependent and uni-directional information flow from the system to the reservoir associated to time-dependent Markovian processes.
Sample size for logistic regression with small response probability
Whittemore, A S
1980-03-01T23:59:59.000Z
The Fisher information matrix for the estimated parameters in a multiple logistic regression can be approximated by the augmented Hessian matrix of the moment generating function for the covariates. The approximation is valid when the probability of response is small. With its use one can obtain a simple closed form estimate of the asymptotic covariance matrix of the maximum likelihood parameter estimates, and thus approximate sample sizes needed to test hypotheses about the parameters. The method is developed for selected distributions of a single covariate, and for a class of exponential-type distributions of several covariates. It is illustrated with an example concerning risk factors for coronary heart disease.
Nonparametric Regression using the Concept of Minimum Energy
Mike Williams
2011-07-12T23:59:59.000Z
It has recently been shown that an unbinned distance-based statistic, the energy, can be used to construct an extremely powerful nonparametric multivariate two sample goodness-of-fit test. An extension to this method that makes it possible to perform nonparametric regression using multiple multivariate data sets is presented in this paper. The technique, which is based on the concept of minimizing the energy of the system, permits determination of parameters of interest without the need for parametric expressions of the parent distributions of the data sets. The application and performance of this new method is discussed in the context of some simple example analyses.
Galtchouk, Leonid
2008-01-01T23:59:59.000Z
An adaptive nonparametric estimation procedure is constructed for the estimation problem of heteroscedastic regression when the noise variance depends on the unknown regression. A non-asymptotic upper bound for a quadratic risk (an oracle inequality) is constructed.
Sharp non-asymptotic oracle inequalities for nonparametric heteroscedastic regression models
Galtchouk, Leonid
2010-01-01T23:59:59.000Z
An adaptive nonparametric estimation procedure is constructed for heteroscedastic regression when the noise variance depends on the unknown regression. A non-asymptotic upper bound for a quadratic risk (oracle inequality) is obtained
Teodorescu, Remus
Aalborg Universitet Photovoltaic Array Condition Monitoring Based on Online Regression of the 39th IEEE Photovoltaic Specialists Conference, PVSC 2013 DOI (link to publication from Publisher): 10., & Teodorescu, R. (2013). Photovoltaic Array Condition Monitoring Based on Online Regression of Performance
Do, Hyunsook
Infrastructure Support for Controlled Experimentation with Software Testing and Regression Testing@cse.unl.edu January 18, 2004 Abstract Where the development, understanding, and assessment of software testing infrastructure to support controlled experimentation with software testing and regression testing techniques
Do, Hyunsook
Infrastructure Support for Controlled Experimentation with Software Testing and Regression Testing@cse.unl.edu April 13, 2005 Abstract Where the development, understanding, and assessment of software testing infrastructure to support controlled experimentation with software testing and regression testing techniques
ECOLOGIC REGRESSION ANALYSIS AND THE STUDY OF THE INFLUENCE OF AIR QUALITY ON MORTALITY
Selvin, S.
2014-01-01T23:59:59.000Z
Orcutt, An empirical analysis of air pollution dose-responseIf ecologic regression analysis of air quality and mortality
Data regression using the TI's calculators TI 82 and TI 83
Torres, Rodolfo
Data regression using the TI's calculators TI 82 and TI 83 TI 85 TI 86 Doing data regression-data in L1 and the y-data in L2. Doing the regression Press to get the statistics menu. Select and select the type of regression. For example, , and press .CALC LinReg ax+b The screen says LinReg; enter (key 2nd 1
Structure based chemical shift prediction using Random Forests non-linear regression
Langmead, Christopher James
Structure based chemical shift prediction using Random Forests non-linear regression K. Arun-ordinates will permit close study of this relationship. This paper presents a novel non-linear regression based ap, regression, Random Forests #12;Abstract Protein nuclear magnetic resonance (NMR) chemical shifts are among
COMPSTAT'2004 Symposium c Physica-Verlag/Springer 2004 ROBUST REGRESSION QUANTILES WITH
to robustly estimate linear regression quantiles with censored data. We adjust the estimator recently points. 1 Introduction We consider the linear regression setting, in which we have to model the re, the deepest regression estimator [5], which has been defined for non censored data. In Section 4 we
Non-Linear Continuum Regression Using Genetic Programming Ben.McKay@ncl.ac.uk
Fernandez, Thomas
Non-Linear Continuum Regression Using Genetic Programming Ben McKay Ben.McKay@ncl.ac.uk Mark Willis In this contribution, genetic programming is combined with continuum regression to produce two novel non-linear-based' strategy. Having discussed continuum regression, the modifications required to extend the algorithm for non-linear
Consistency of the posterior distribution and MLE for piecewise linear regression
Paris-Sud XI, Université de
Consistency of the posterior distribution and MLE for piecewise linear regression Tristan Launay1 and that of the Bayes estimator for a two-phase piecewise linear regression mdoel where the break-point is unknown and be the unknown regression coefficient of the non-zero phase. The observations X1:n = (X1, . . . , Xn) depend
Statistical prediction of aircraft trajectory: regression methods vs point-mass model
Paris-Sud XI, Université de
the altitude of climbing aircraft. In addition to the standard linear regression model, two common non-linear, BADA, linear regression, neural networks, Loess. INTRODUCTION Predicting aircraft trajectoriesStatistical prediction of aircraft trajectory: regression methods vs point-mass model M. Ghasemi
Higher-Order Partial Least Squares (HOPLS): A Generalized Multi-Linear Regression Method
Cichocki, Andrzej
1 Higher-Order Partial Least Squares (HOPLS): A Generalized Multi-Linear Regression Method Qibin Regression (PLSR) - a multivariate method which, in contrast to Multiple Linear Regression (MLR. A standard way to optimize the model parameters is the Non- linear Iterative Partial Least Squares (NIPALS
Functional Coefficient Regression Models for Non-linear Time Series: A Polynomial
Shen, Haipeng
Functional Coefficient Regression Models for Non-linear Time Series: A Polynomial Spline Approach of functional coefficient regression models for non-linear time series. Consistency and rate of convergence regression model extends several familiar non-linear time series models such as the exponential
Logistic Regression and Bayesian Model Selection in Estimation of Probability of Success
Shemyakin, Arkady
1 1 Logistic Regression and Bayesian Model Selection in Estimation of Probability of Success Arkady ABSTRACT Logistic regression and linear discriminant analysis are used to estimate probability of success X is analyzed as an explanatory variable. A comparison is made between logistic regression technique
Logistic Regression Advanced Methods for Data Analysis (36-402/36-608)
Tibshirani, Ryan
Logistic Regression Advanced Methods for Data Analysis (36-402/36-608) Spring 2014 1 Classification. In this lecture we will learn one of the most common tools: logistic regression. You should know function) 1.2 Why not just use least squares? · Before we present logistic regression, we address
Feature Selection for Support Vector Regression in the Application of Building Energy Prediction
Paris-Sud XI, Université de
Feature Selection for Support Vector Regression in the Application of Building Energy Prediction--When using support vector regression to predict building energy consumption, since the energy influence and reduces the computational time. Keywords-support vector regression; feature selection; build- ing; energy
IBLStreams: A System for Instance-Based Classification and Regression on Data Streams
Hüllermeier, Eyke
IBLStreams: A System for Instance-Based Classification and Regression on Data Streams Ammar Shaker to classification and regression problems. In comparison to model-based methods for learning on data streams. Keywords: Data streams, classification, regression, instance-based learn- ing, concept drift. 1 #12
On Test Suite Composition and Cost-Effective Regression Testing Gregg Rothermel
Rothermel, Gregg
On Test Suite Composition and Cost-Effective Regression Testing Gregg Rothermel , Sebastian Elbaum}@cse.unl.edu August 31, 2004 Abstract Regression testing is an expensive testing process used to re-validate software as it evolves. Various methodologies for improving regression testing processes have been explored, but the cost
On Test Suite Composition and Cost-Effective Regression Testing. Gregg Rothermel
Rothermel, Gregg
On Test Suite Composition and Cost-Effective Regression Testing. Gregg Rothermel , Sebastian Elbaum}@cse.unl.edu August 30, 2003 Abstract Regression testing is an expensive testing process used to re-validate software as it evolves. Various methodologies for improving regression testing processes have been explored, but the cost
Logistic Regression and Artificial Neural Networks for Classification of Ovarian Tumors
Logistic Regression and Artificial Neural Networks for Classification of Ovarian Tumors C. Lu1 , J to generate and evaluate both logistic regression models and artificial neural network (ANN) models to predict, including explorative univariate and multivariate analysis, and the development of the logistic regression
A regression model with a hidden logistic process for signal parametrization
Chamroukhi, Faicel
A regression model with a hidden logistic process for signal parametrization F. Chamroukhi, A. Samé (INRETS/UTC-France) ESANN 2009 April 24 2009 1 / 21 #12;Outline 1 Context 2 The piecewise regression approach 3 The proposed regression approach 4 Parameter estimation 5 Experiments Faicel Chamroukhi (INRETS
Modeling Animal-Vehicle Collisions Using Diagonal Inflated Bivariate Poisson Regression
Washington at Seattle, University of
1 Modeling Animal-Vehicle Collisions Using Diagonal Inflated Bivariate Poisson Regression of highway AVCs, this study adopts a diagonal inflated bivariate Poisson regression method, an inflated version of bivariate Poisson regression model, to fit the reported AVC and carcass removal data sets
Regression Model for Daily Maximum Stream Temperature David W. Neumann1
Balaji, Rajagopalan
Regression Model for Daily Maximum Stream Temperature David W. Neumann1 ; Balaji Rajagopalan2 for the summer period. The model is created using a stepwise linear regression procedure to select significant-9372 2003 129:7 667 CE Database subject headings: Decision support systems; Regression models; California
Statistica Sinica 19 (2009), 801-817 VARIABLE SELECTION IN QUANTILE REGRESSION
Liu, Yufeng
2009-01-01T23:59:59.000Z
Statistica Sinica 19 (2009), 801-817 VARIABLE SELECTION IN QUANTILE REGRESSION Yichao Wu and Yufeng inception in Koenker and Bassett (1978), quantile regression has become an important and widely used- lection aspect of penalized quantile regression. Under some mild conditions, we demonstrate the oracle
Regression of Multicomponent Sticking Probabilities Using a Genetic Algorithm Ian J. Laurenzi*
Regression of Multicomponent Sticking Probabilities Using a Genetic Algorithm Ian J. LaurenziVania, Philadelphia, PennsylVania 19104 A genetic algorithm (GA) was developed for the purpose of regressing processes were then simulated under physiological conditions via Monte Carlo. The GA successfully regressed
Estimating Litter Decomposition Rate in Single-Pool Models Using Nonlinear Beta Regression
Thomas, David D.
Estimating Litter Decomposition Rate in Single-Pool Models Using Nonlinear Beta Regression Etienne the performance of nonlinear regression using the beta distribution, which is well-suited to bounded data and this type of heteroscedasticity, to standard nonlinear regression (normal errors) on simulated and real
REGRESSION PERFORMANCE OF GROUP LASSO FOR ARBITRARY DESIGN MATRICES Marco F. Duarte,1,
Bajwa, Waheed U.
REGRESSION PERFORMANCE OF GROUP LASSO FOR ARBITRARY DESIGN MATRICES Marco F. Duarte,1, Waheed U.duarte,w.bajwa,robert.calderbank}@duke.edu ABSTRACT In many linear regression problems, explanatory variables are activated in groups or clusters; group lasso has been proposed for regression in such cases. This paper studies the non- asymptotic
Autologistic Regression Analysis of Spatial-Temporal Binary Data via Monte Carlo
Aukema, Brian
Autologistic Regression Analysis of Spatial-Temporal Binary Data via Monte Carlo Maximum Likelihood regression analysis of binary data that are measured on a spatial lattice and repeatedly over discrete time points. We propose a spatial- temporal autologistic regression model and draw statistical inference via
Kutas, Marta
Regression-based estimation of ERP waveforms: II. Nonlinear effects, overlap correction regression-based approach to estimating ERP waveforms. Here, we build on this foundation, showing how rERP framework provides a flexible way to adapt well-known regression techniques to the problem of estimating
Crawford, John R.
Using Regression Equations Built From Summary Data in the Neuropsychological Assessment of the Individual Case John R. Crawford University of Aberdeen Paul H. Garthwaite The Open University Regression that there is a large reservoir of published data that could be used to build regression equations; these equations
Regression Analysis Using Weighted Least Squares Robert Drucker, University of Washington (May 1996)
Washington at Seattle, University of
Regression Analysis Using Weighted Least Squares Robert Drucker, University of Washington (May 1996) 1. Theory The jth order leastsquares regression problem can be stated in matrix notation as Y = Xb regression analysis assumes that the residuals are independent identically dis tributed (iid) random
Estimating Salinity between 25 and 45S in the Atlantic Ocean Using Local Regression
Estimating Salinity between 25° and 45°S in the Atlantic Ocean Using Local Regression W. C. THACKER and temperature in the South Atlantic is quantified with the aid of local regression. To capture the spatial fitting regression models to the profile data is considerably more work than using published clima
Multiple regression on distance matrices: a multivariate spatial analysis tool Jeremy W. Lichstein
Lichstein, Jeremy W.
Multiple regression on distance matrices: a multivariate spatial analysis tool Jeremy W. Lichstein, Spatial autocorrelation Abstract I explore the use of multiple regression on distance matrices (MRM regression of a response matrix on any number of explanatory matrices, where each matrix contains distances
Efficient Regression of General-Activity Human Poses from Depth Images Ross Girshick
Kohli, Pushmeet
Efficient Regression of General-Activity Human Poses from Depth Images Ross Girshick Jamie Shotton of several decision-tree training ob- jectives. Key aspects of our work include: regression di- rectly from the regression-based ap- proaches that have been a staple of monocular 2D human pose estimation [1, 19, 10, 15
Practical Selection of SVM Parameters and Noise Estimation for SVM Regression
Cherkassky, Vladimir
1 Practical Selection of SVM Parameters and Noise Estimation for SVM Regression Vladimir Cherkassky 55455, USA Abstract We investigate practical selection of meta-parameters for SVM regression (that is using several low-dimensional and high-dimensional regression problems. Further, we point out
Adaptive Regression Testing Strategy: An Empirical Study Md. Junaid Arafeen and Hyunsook Do
Do, Hyunsook
Adaptive Regression Testing Strategy: An Empirical Study Md. Junaid Arafeen and Hyunsook Do in different versions. These factors can affect the costs and benefits of regression testing techniques in different ways, and thus, there may be no single regression testing technique that is the most cost
Regression Models with Interval Censoring Jian Huang and Jon A. Wellner 1
Wellner, Jon A.
Regression Models with Interval Censoring Jian Huang and Jon A. Wellner 1 University of Washington October 6, 1993 Abstract In this paper we discuss estimation in semiparametric regression models with interval censoring, with emphasis on estimation of the regression parameter . The first section surveys
Spatial autocorrelation approaches to testing residuals from least squares regression
Chen, Yanguang
2015-01-01T23:59:59.000Z
In statistics, the Durbin-Watson test is always employed to detect the presence of serial correlation of residuals from a least squares regression analysis. However, the Durbin-Watson statistic is only suitable for ordered time or spatial series. If the variables comprise cross-sectional data coming from spatial random sampling, the Durbin-Watson will be ineffectual because the value of Durbin-Watson's statistic depends on the sequences of data point arrangement. Based on the ideas from spatial autocorrelation, this paper presents two new statistics for testing serial correlation of residuals from least squares regression based on spatial samples. By analogy with the new form of Moran's index, an autocorrelation coefficient is defined with a standardized residual vector and a normalized spatial weight matrix. Then on the analogy of the Durbin-Watson statistic, a serial correlation index is constructed. As a case, the two statistics are applied to the spatial sample of 29 China's regions. These results show th...
arXiv:1412.2844v1[stat.CO]9Dec2014 Optimal Reduced Isotonic Regression
Stout, Quentin F.
arXiv:1412.2844v1[stat.CO]9Dec2014 Optimal Reduced Isotonic Regression Janis Hardwick and Quentin F regression is a shape-constrained nonparametric regression in which the regression is an increasing step function. For n data points, the number of steps in the isotonic regression may be as large as n
Fast transform from an adaptive multi-wavelet representation to a partial Fourier representation
Jia, Jun [ORNL; Harrison, Robert J [ORNL; Fann, George I [ORNL
2010-01-01T23:59:59.000Z
We present a fast algorithm to compute the partial transformation of a function represented in an adaptive pseudo-spectral multi-wavelet representation to a partial Fourier representation. Such fast transformations are useful in many contexts in physics and engineering, where changes of representation from a piece wise polynomial basis to a Fourier basis. The algorithm is demonstrated for a Gaussian in one and in three dimensions. For 2D, we apply this approach to a Gaussian in a periodic domain. The accuracy and the performance of this method is compared with direct summation.
Solving three-body scattering problem in the momentum lattice representation
V. N. Pomerantsev; V. I. Kukulin; O. A. Rubtsova
2008-12-02T23:59:59.000Z
A brief description of the novel approach towards solving few-body scattering problems in a finite-dimensional functional space of the $L_2$-type is presented. The method is based on the complete few-body continuum discretization in the basis of stationary wave packets. This basis, being transformed to the momentum representation, leads to the cell-lattice-like discretization of the momentum space. So the initial scattering problem can be formulated on the multi-dimensional momentum lattice which makes it possible to reduce the solution of any scattering problem above the breakup threshold (where the integral kernels include, in general, some complicated moving singularities) to convenient simple matrix equations which can be solved on the real energy axis. The phase shifts and inelasticity parameters for the three-body $nd$ elastic scattering with MT I-III $NN$ potential both below and above the three-body breakup threshold calculated with the proposed wave-packet technique are in a very good agreement with the previous accurate benchmark calculation results.
FROM NORLUND MATRICES TO LAPLACE REPRESENTATIONS
Sinnamon, Gordon J.
(and not too large) on the line Re z = 0, the Laplace transform LF is just the (Poisson extension of the) Fourier transform of F. It is therefore appropriate to view the power series representation¨orlund matrices and corresponding convolution operators on the line. Analogous inequalities are proved for power
THE BREUILMEZARD CONJECTURE FOR POTENTIALLY BARSOTTITATE REPRESENTATIONS.
Kisin, Mark
THE BREUILM´EZARD CONJECTURE FOR POTENTIALLY BARSOTTITATE REPRESENTATIONS. TOBY GEE AND MARK, proving a variety of results including the BuzzardDiamondJarvis conjecture. Contents Overview. 2.3. Patching 14 3.4. Potential diagonalizability 18 3.5. Local results 20 4. The BuzzardDiamondJarvis
Galois representations with quaternion multiplication associated to ...
A.O.L. Atkin; Wen-Ching Winnie Li; Tong Liu; Ling Long
2013-09-18T23:59:59.000Z
Aug 19, 2013 ... Modularity of Scholl representations when d = 1. 6228. 4.2. ... 1. Introduction. To a d-dimensional space S?(?) of cusp forms of weight ? > 2 for a noncongru- ence subgroup ? ...... MR2038777 (2004m:11089). [BLTDR10] T.
Radon Transform Inversion using the Shearlet Representation
Labate, Demetrio
Radon Transform Inversion using the Shearlet Representation Flavia Colonna Department The inversion of the Radon transform is a classical ill-posed inverse problem where some method-optimal rate of convergence in estimating a large class of images from noisy Radon data. This is achieved
ACQUIRED EQUIVALENCE CHANGES STIMULUS REPRESENTATIONS , D. SHOHAMY
Shohamy, Daphna
ACQUIRED EQUIVALENCE CHANGES STIMULUS REPRESENTATIONS M. MEETER 1 , D. SHOHAMY 2 , AND C.E. MYERS 3 UNIVERSITY 3 DEPT. OF PSYCHOLOGY, RUTGERS UNIVERSITY Acquired equivalence is a paradigm in which of feature salience. A different way of conceptualizing acquired equivalence is in terms of strategic
Shape Recipes: Scene Representations that Refer
Freeman, William T.
Shape Recipes: Scene Representations that Refer to the Image William T. Freeman and Antonio to estimate and store. We propose a low-dimensional rep- resentation, called a scene recipe, that relies on the image itself to de- scribe the complex scene configurations. Shape recipes are an example
Adaptive Representation of Specular Light Flux
Montréal, Université de
Adaptive Representation of Specular Light Flux Normand Bri`ere Pierre Poulin D´epartement d in all but the simplest con- figurations. To capture their appearance, we present an adaptive approach based upon light beams. The coher- ence between light rays forming a light beam greatly re- duces
Memory Space Representation Heterogeneous Network Process Migration
Sun, Xian-He
Memory Space Representation for Heterogeneous Network Process Migration Kasidit Chanchio Xian@bit.csc.lsu.edu http://www.csc.lsu.edu/~scs/ Abstract A major difficulty of heterogeneous process migration is how and effective for heterogeneous network process migration. 1. Introduction As network computing becomes
Bayes Nets Representation: joint distribution and conditional
Mitchell, Tom
Bayes Nets Representation: joint distribution and conditional independence Yi Zhang 10-701, Machine joint distribution of BNs Infer C. I. from factored joint distributions D-separation (motivation) 2 structure All about the joint distribution of variables ! Conditional independence assumptions are useful
Taylor Expansion Diagrams: A Canonical Representation for
Kalla, Priyank
Taylor series expansion that allows one to model word-level signals as algebraic symbols. This power systems has made it essential to address verification issues at early stages of the design cycle representations. TEDs are applicable to modeling, symbolic simulation, and equivalence verification of dataflow
Zander, Jessica Selene
2012-01-01T23:59:59.000Z
Narratives of Contamination: Representations of Race,Fall 2012 Narratives of Contamination: Representations ofAbstract Narratives of Contamination: Representations of
Hall, Sharon J.
Figure 3. Socioeconomics drive biomass too. Simple regression with untrans- formed variables. Solid line represents the estimated regression line, whereas the dashed lines represent the 95% confidence metropolitan area. I hypothesized that income is the driving factor of vegetation coverage, primarily affecting
Representation of Energy Use in the Food Products Industry
Elliott, N. R.
2007-01-01T23:59:59.000Z
Traditional representations of energy in the manufacturing sector have tended to represent energy end-uses rather than actual energy service demands. While this representation if quite adequate for understanding how energy is used today...
Representation of Energy Use in the Food Products Industry
Elliott, N. R.
2007-01-01T23:59:59.000Z
Traditional representations of energy in the manufacturing sector have tended to represent energy end-uses rather than actual energy service demands. While this representation if quite adequate for understanding how energy is used today...
Lexical semantics and knowledge representation in multilingual sentence generation
Toronto, University of
Lexical semantics and knowledge representation in multilingual sentence generation by Manfred Stede semantics and knowledge representation in multilingual sentence generation Manfred Stede Doctor approach to automatic language generation that focuses on the need to produce a range of di erent
A Visual Analytics Approach for Correlation, Classification, and Regression Analysis
Steed, Chad A [ORNL; SwanII, J. Edward [Mississippi State University (MSU); Fitzpatrick, Patrick J. [Mississippi State University (MSU); Jankun-Kelly, T.J. [Mississippi State University (MSU)
2012-02-01T23:59:59.000Z
New approaches that combine the strengths of humans and machines are necessary to equip analysts with the proper tools for exploring today's increasing complex, multivariate data sets. In this paper, a novel visual data mining framework, called the Multidimensional Data eXplorer (MDX), is described that addresses the challenges of today's data by combining automated statistical analytics with a highly interactive parallel coordinates based canvas. In addition to several intuitive interaction capabilities, this framework offers a rich set of graphical statistical indicators, interactive regression analysis, visual correlation mining, automated axis arrangements and filtering, and data classification techniques. The current work provides a detailed description of the system as well as a discussion of key design aspects and critical feedback from domain experts.
Robust linear regression with broad distributions of errors
Postnikov, Eugene B
2015-01-01T23:59:59.000Z
We consider the problem of linear fitting of noisy data in the case of broad (say $\\alpha$-stable) distributions of random impacts ("noise"), which can lack even the first moment. This situation, common in statistical physics of small systems, in Earth sciences, in network science or in econophysics, does not allow for application of conventional Gaussian maximum-likelihood estimators resulting in usual least-squares fits. Such fits lead to large deviations of fitted parameters from their true values due to the presence of outliers. The approaches discussed here aim onto the minimization of the width of the distribution of residua. The corresponding width of the distribution can either be defined via the interquantile distance of the corresponding distributions or via the scale parameter in its characteristic function. The methods provide the robust regression even in the case of short samples with large outliers, and are equivalent to the normal least squares fit for the Gaussian noises. Our discussion is il...
A Visual Analytics Approach for Correlation, Classification, and Regression Analysis
Steed, Chad A [ORNL; SwanII, J. Edward [Mississippi State University (MSU); Fitzpatrick, Patrick J. [Mississippi State University (MSU); Jankun-Kelly, T.J. [Mississippi State University (MSU)
2013-01-01T23:59:59.000Z
New approaches that combine the strengths of humans and machines are necessary to equip analysts with the proper tools for exploring today s increasing complex, multivariate data sets. In this paper, a visual data mining framework, called the Multidimensional Data eXplorer (MDX), is described that addresses the challenges of today s data by combining automated statistical analytics with a highly interactive parallel coordinates based canvas. In addition to several intuitive interaction capabilities, this framework offers a rich set of graphical statistical indicators, interactive regression analysis, visual correlation mining, automated axis arrangements and filtering, and data classification techniques. This chapter provides a detailed description of the system as well as a discussion of key design aspects and critical feedback from domain experts.
TURBULENCE-INDUCED RELATIVE VELOCITY OF DUST PARTICLES. IV. THE COLLISION KERNEL
Pan, Liubin [Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138 (United States); Padoan, Paolo, E-mail: lpan@cfa.harvard.edu, E-mail: ppadoan@icc.ub.edu [ICREA and Institut de Cičncies del Cosmos, Universitat de Barcelona, IEEC-UB, Martí Franqučs 1, E-08028 Barcelona (Spain)
2014-12-20T23:59:59.000Z
Motivated by its importance for modeling dust particle growth in protoplanetary disks, we study turbulence-induced collision statistics of inertial particles as a function of the particle friction time, ?{sub p}. We show that turbulent clustering significantly enhances the collision rate for particles of similar sizes with ?{sub p} corresponding to the inertial range of the flow. If the friction time, ?{sub p,} {sub h}, of the larger particle is in the inertial range, the collision kernel per unit cross section increases with increasing friction time, ?{sub p,} {sub l}, of the smaller particle and reaches the maximum at ?{sub p,} {sub l} = ?{sub p,} {sub h}, where the clustering effect peaks. This feature is not captured by the commonly used kernel formula, which neglects the effect of clustering. We argue that turbulent clustering helps alleviate the bouncing barrier problem for planetesimal formation. We also investigate the collision velocity statistics using a collision-rate weighting factor to account for higher collision frequency for particle pairs with larger relative velocity. For ?{sub p,} {sub h} in the inertial range, the rms relative velocity with collision-rate weighting is found to be invariant with ?{sub p,} {sub l} and scales with ?{sub p,} {sub h} roughly as ? ?{sub p,h}{sup 1/2}. The weighting factor favors collisions with larger relative velocity, and including it leads to more destructive and less sticking collisions. We compare two collision kernel formulations based on spherical and cylindrical geometries. The two formulations give consistent results for the collision rate and the collision-rate weighted statistics, except that the spherical formulation predicts more head-on collisions than the cylindrical formulation.
The effect of artificial drying temperature on the quality of early harvested pecan kernels
McLean, Roy William
1988-01-01T23:59:59.000Z
process was complete, the pecans were shelled, sealed in 8303 cans under 27inHg vacuum, and stored at 0 F until analyses were performed. Sixty whole kernels were randomly selected from each sample, objectively evaluated for color and then cold pressed... harvesting entails shaking the tree after the husks have split and allowing them to dry naturally. Once the nuts 10 were dried, they were shelled and sealed in ()303 cans, sealed under vacuum and stored at 0 F until analyses were performed. Methods...
Anomaly-free representations of the holonomy-flux algebra
SangChul Yoon
2008-09-07T23:59:59.000Z
We work on the uniqueness, gr-qc/0504147, of representations of the holonomy-flux algebra in loop quantum gravity. We argue that for analytic diffeomorphisms, the flux operators can be only constants as functions on the configuration space in representations with no anomaly, which are zero in the standard representation.
Representation Theory, Geometry & Combinatorics Organizer: M. Haiman and N. Reshetikhin
Haiman, Mark D.
Representation Theory, Geometry & Combinatorics Seminar Organizer: M. Haiman and N. Reshetikhin course: Representation theory and the X-ray transform The X-ray transform (also called the Funk transform tools from complex analysis and the representation theory of Lie groups. Lecture 1: Differential
Support vector methods for survival analysis: a comparison between ranking and regression
techniques for the estimation of non-linear transformation models for the analysis of survival data. Methods is the use of non-linear kernels im- plementing automatically non-parametric effects of the covariates the advantage that they are easily extendable towards non-linear models without the need to check non-linearities
Huang, Yi-Zhi
Quantum Hall systems Representation theory of vertex operator algebras Applications The end Quantum;Quantum Hall systems Representation theory of vertex operator algebras Applications The end Outline 1 An approach to a fundamental conjecture #12;Quantum Hall systems Representation theory of vertex operator
Huang, Yi-Zhi
Quantum Hall systems Representation theory of vertex operator algebras Applications The end Quantum Science, CAS #12;Quantum Hall systems Representation theory of vertex operator algebras Applications to a fundamental conjecture #12;Quantum Hall systems Representation theory of vertex operator algebras Applications
Animal representations and animal remains at Çatalhöyük
Russell, Nerissa; Meece, Stephanie
2006-01-01T23:59:59.000Z
(Level VII). Volcano above town plan, leopard skin above geometric design, or other representations? Level VI paintings lack fully convinc ing animal depictions. A patch of painting on the east wall of building VIA.66 includes a number of geomet ric... the centrepieces of the north walls of two rather similar buildings. In a sense they parallel the situation in the faunal assemblage, where cattle are not terribly common, but figure prominently in cer emonial consumption (see Russell & Martin, Volume 4...
Ford, Bryan
The Flux OSKit: A Substrate for Kernel and Language Research Bryan Ford Godmar Back Greg Benson Jay a basic useful OS core--e.g., the functionality traditionally found in the Unix kernel--entirely from suited for physical memory and its Ford, Back, and Lepreau are at the Univ. of Utah (baford
Brown, Angela Demke
The Flux OSKiti A Substrate for Kernel and Language Research Bryan Ford Godmar Back Greg Benson Jay a basic useful OS core-eg., the functionality traditionally found in the Unix kernel-entirely from scratch for physical memory and its Ford, Back, and Lqreau are at the Univ. of Utah @aford,gback,lepreau- @cs
Virasoro Representations on (Diff S1)/S1 Coadjoint Orbits
Washington Taylor IV
1992-04-28T23:59:59.000Z
A new set of realizations of the Virasoro algebra on a bosonic Fock space are found by explicitly computing the Virasoro representations associated with coadjoint orbits of the form (Diff S1) / S1. Some progress is made in understanding the unitary structure of these representations. The characters of these representations are exactly the bosonic partition functions calculated previously by Witten using perturbative and fixed-point methods. The representations corresponding to the discrete series of unitary Virasoro representations with c <= 1 are found to be reducible in this formulation, confirming a conjecture by Aldaya and Navarro-Salas.
Sérgio Szpigel; Varese S. Timóteo
2012-07-26T23:59:59.000Z
We apply the subtracted kernel method (SKM), a renormalization approach based on recursive multiple subtractions performed in the kernel of the scattering equation, to the chiral nucleon-nucleon (NN) interactions up to next-to-next-to-leading-order (NNLO). We evaluate the phase-shifts in the 1S0 channel at each order in Weinberg's power counting scheme and in a modified power counting scheme which yields a systematic power-law improvement. We also explicitly demonstrate that the SKM procedure is renormalization group invariant under the change of the subtraction scale through a non-relativistic Callan-Symanzik flow equation for the evolution of the renormalized NN interactions.
GPU Kernels for High-Speed 4-Bit Astrophysical Data Processing
Klages, Peter; Denman, Nolan; Recnik, Andre; Sievers, Jonathan; Vanderlinde, Keith
2015-01-01T23:59:59.000Z
Interferometric radio telescopes often rely on computationally expensive O(N^2) correlation calculations; fortunately these computations map well to massively parallel accelerators such as low-cost GPUs. This paper describes the OpenCL kernels developed for the GPU based X-engine of a new hybrid FX correlator. Channelized data from the F-engine is supplied to the GPUs as 4-bit, offset-encoded real and imaginary integers. Because of the low bit width of the data, two values may be packed into a 32-bit register, allowing multiplication and addition of more than one value with a single fused multiply-add instruction. With this data and calculation packing scheme, as many as 5.6 effective tera-operations per second (TOPS) can be executed on a 4.3 TOPS GPU. The kernel design allows correlations to scale to large numbers of input elements, limited only by maximum buffer sizes on the GPU. This code is currently working on-sky with the CHIME Pathfinder Correlator in BC, Canada.
Liu, Derek, E-mail: dmliu@ualberta.ca; Sloboda, Ron S. [Department of Medical Physics, Cross Cancer Institute, Edmonton, Alberta T6G 1Z2, Canada and Department of Oncology, University of Alberta, Edmonton, Alberta T6G 2R3 (Canada)] [Department of Medical Physics, Cross Cancer Institute, Edmonton, Alberta T6G 1Z2, Canada and Department of Oncology, University of Alberta, Edmonton, Alberta T6G 2R3 (Canada)
2014-05-15T23:59:59.000Z
Purpose: Boyer and Mok proposed a fast calculation method employing the Fourier transform (FT), for which calculation time is independent of the number of seeds but seed placement is restricted to calculation grid points. Here an interpolation method is described enabling unrestricted seed placement while preserving the computational efficiency of the original method. Methods: The Iodine-125 seed dose kernel was sampled and selected values were modified to optimize interpolation accuracy for clinically relevant doses. For each seed, the kernel was shifted to the nearest grid point via convolution with a unit impulse, implemented in the Fourier domain. The remaining fractional shift was performed using a piecewise third-order Lagrange filter. Results: Implementation of the interpolation method greatly improved FT-based dose calculation accuracy. The dose distribution was accurate to within 2% beyond 3 mm from each seed. Isodose contours were indistinguishable from explicit TG-43 calculation. Dose-volume metric errors were negligible. Computation time for the FT interpolation method was essentially the same as Boyer's method. Conclusions: A FT interpolation method for permanent prostate brachytherapy TG-43 dose calculation was developed which expands upon Boyer's original method and enables unrestricted seed placement. The proposed method substantially improves the clinically relevant dose accuracy with negligible additional computation cost, preserving the efficiency of the original method.
Density-Aware Clustering Based on Aggregated Heat Kernel and Its Transformation
DOE Public Access Gateway for Energy & Science Beta (PAGES Beta)
Huang, Hao; Yoo, Shinjae; Yu, Dantong; Qin, Hong
2015-06-01T23:59:59.000Z
Current spectral clustering algorithms suffer from the sensitivity to existing noise, and parameter scaling, and may not be aware of different density distributions across clusters. If these problems are left untreated, the consequent clustering results cannot accurately represent true data patterns, in particular, for complex real world datasets with heterogeneous densities. This paper aims to solve these problems by proposing a diffusion-based Aggregated Heat Kernel (AHK) to improve the clustering stability, and a Local Density Affinity Transformation (LDAT) to correct the bias originating from different cluster densities. AHK statistically\\ models the heat diffusion traces along the entire time scale, somore »it ensures robustness during clustering process, while LDAT probabilistically reveals local density of each instance and suppresses the local density bias in the affinity matrix. Our proposed framework integrates these two techniques systematically. As a result, not only does it provide an advanced noise-resisting and density-aware spectral mapping to the original dataset, but also demonstrates the stability during the processing of tuning the scaling parameter (which usually controls the range of neighborhood). Furthermore, our framework works well with the majority of similarity kernels, which ensures its applicability to many types of data and problem domains. The systematic experiments on different applications show that our proposed algorithms outperform state-of-the-art clustering algorithms for the data with heterogeneous density distributions, and achieve robust clustering performance with respect to tuning the scaling parameter and handling various levels and types of noise.« less
Validi, AbdoulAhad, E-mail: validiab@msu.edu
2014-03-01T23:59:59.000Z
This study introduces a non-intrusive approach in the context of low-rank separated representation to construct a surrogate of high-dimensional stochastic functions, e.g., PDEs/ODEs, in order to decrease the computational cost of Markov Chain Monte Carlo simulations in Bayesian inference. The surrogate model is constructed via a regularized alternative least-square regression with Tikhonov regularization using a roughening matrix computing the gradient of the solution, in conjunction with a perturbation-based error indicator to detect optimal model complexities. The model approximates a vector of a continuous solution at discrete values of a physical variable. The required number of random realizations to achieve a successful approximation linearly depends on the function dimensionality. The computational cost of the model construction is quadratic in the number of random inputs, which potentially tackles the curse of dimensionality in high-dimensional stochastic functions. Furthermore, this vector-valued separated representation-based model, in comparison to the available scalar-valued case, leads to a significant reduction in the cost of approximation by an order of magnitude equal to the vector size. The performance of the method is studied through its application to three numerical examples including a 41-dimensional elliptic PDE and a 21-dimensional cavity flow.
Real-time semiparametric regression BY J. LUTS1, T. BRODERICK2 AND M.P. WAND1
Wand, Matt
regression refers to a large class of regression models that provide for non-linear predictor effects using regression is quite broad and includes, as special cases, generalized linear mixed models, generalizedReal-time semiparametric regression BY J. LUTS1, T. BRODERICK2 AND M.P. WAND1 1 School
Regression-based estimates of the rate of accumulation of anthropogenic CO2 in the ocean: A fresh February 2012 Available online 23 February 2012 Keywords: Carbon dioxide Regression MLR eMLR Regression and guidelines for improvement are presented. Following these guidelines leads to a local two- regression method
Nicholas G Phillips; B. L. Hu
2002-09-17T23:59:59.000Z
In Paper II [N. G. Phillips and B. L. Hu, previous abstract] we presented the details for the regularization of the noise kernel of a quantum scalar field in optical spacetimes by the modified point separation scheme, and a Gaussian approximation for the Green function. We worked out the regularized noise kernel for two examples: hot flat space and optical Schwarzschild metric. In this paper we consider noise kernels for a scalar field in the Schwarzschild black hole. Much of the work in the point separation approach is to determine how the divergent piece conformally transforms. For the Schwarzschild metric we find that the fluctuations of the stress tensor of the Hawking flux in the far field region checks with the analytic results given by Campos and Hu earlier [A. Campos and B. L. Hu, Phys. Rev. D {\\bf 58} (1998) 125021; Int. J. Theor. Phys. {\\bf 38} (1999) 1253]. We also verify Page's result [D. N. Page, Phys. Rev. {\\bf D25}, 1499 (1982)] for the stress tensor, which, though used often, still lacks a rigorous proof, as in his original work the direct use of the conformal transformation was circumvented. However, as in the optical case, we show that the Gaussian approximation applied to the Green function produces significant error in the noise kernel on the Schwarzschild horizon. As before we identify the failure as occurring at the fourth covariant derivative order.
Growth of Hereford-Kedah Kelantan calves fed oil palm fronds and palm kernel cake based diet
Paris-Sud XI, Université de
million hectares of land under oil palm cultivation. The palm oil mills yield a number of by-products, the important by- product is the oil palm frond (OPF) which can be utilised fresh or ensiled. HerefordGrowth of Hereford-Kedah Kelantan calves fed oil palm fronds and palm kernel cake based diet I
Regression analysis of technical parameters affecting nuclear power plant performances
Ghazy, R.; Ricotti, M. E.; Trueco, P. [Politecnico di Milano, Via La Masa, 34, 20156 Milano (Italy)
2012-07-01T23:59:59.000Z
Since the 80's many studies have been conducted in order to explicate good and bad performances of commercial nuclear power plants (NPPs), but yet no defined correlation has been found out to be totally representative of plant operational experience. In early works, data availability and the number of operating power stations were both limited; therefore, results showed that specific technical characteristics of NPPs were supposed to be the main causal factors for successful plant operation. Although these aspects keep on assuming a significant role, later studies and observations showed that other factors concerning management and organization of the plant could instead be predominant comparing utilities operational and economic results. Utility quality, in a word, can be used to summarize all the managerial and operational aspects that seem to be effective in determining plant performance. In this paper operational data of a consistent sample of commercial nuclear power stations, out of the total 433 operating NPPs, are analyzed, mainly focusing on the last decade operational experience. The sample consists of PWR and BWR technology, operated by utilities located in different countries, including U.S. (Japan)) (France)) (Germany)) and Finland. Multivariate regression is performed using Unit Capability Factor (UCF) as the dependent variable; this factor reflects indeed the effectiveness of plant programs and practices in maximizing the available electrical generation and consequently provides an overall indication of how well plants are operated and maintained. Aspects that may not be real causal factors but which can have a consistent impact on the UCF, as technology design, supplier, size and age, are included in the analysis as independent variables. (authors)
Representations of groups of order 16
McCarthy, Edmond Robert
1966-01-01T23:59:59.000Z
, 16) K - {10, 14) The analysis al. so shows that G contains three subgroups H. of order 2; only one of which is normal. This is the i subgroup consisting of the elements (1, 5}. The factor group G/H is isomorphic to C 4 C2. Since the Cayley table..., 13 10, 14 11, 15 37 With this correspondence established we need only refer back to Table II, the character table of C 4 C2, to begin writing out representations of Group Six. For example, in Table II we find D&(3) = -1. If TK...
Representations of the groups of order 24
Strange, John Billy
1967-01-01T23:59:59.000Z
: MATHEMATICS REPRESENTATIONS OF THE GROUPS OF ORDER 24 A Thesis By JOHN BILLY STRANGE Approved as to style and content by: I tl ~Chairman of Commrtteeg YMe ber C o QH d I D p ~t'Dt May l967 ACKNOWLEDGMENT During the past eighteen months it has been... cyclic group of order i. 1 The remaining groups with symbols are listed below. 10. 12 (Dihedral of order 24) 11. 12. Q] 2 13. (4, 2 l 2, 2) (Symmetric of order 24) (Quaternion of order 24) 14. &-2, 2, 3& 15. &2, 3, 3& 10 A recent discovery by D...
BASHIR et al.: GAIT REPRESENTATION USING FLOW FIELDS 1 Gait Representation Using Flow Fields
Gong, Shaogang
the human body configuration (e.g. 2D/3D skeletons) and the model parameters estimated over time encode approaches such as Gait Energy Image (GEI) and Motion Silhouettes Image (MSI) capture only the motion inten unchanged freely in print or electronic forms. #12;2 BASHIR et al.: GAIT REPRESENTATION USING FLOW FIELDS
Reisslein, Martin
Teaching With Concrete and Abstract Visual Representations: Effects on Students' Problem Solving/or abstract visual problem representa- tions during instruction on students' problem-solving practice, near outperformed Groups A and C on problem-solving practice in Experiments 1 and 2 and outperformed Group C
Combining Regression Trees and Radial Basis Function Networks Mark Orr, John Hallam,
Edinburgh, University of
a model using linear regression. The non-linear transformation is controlled by a set of m basis functions, 1988] transform the n- dimensional inputs non-linearly to an m-dimensional space and then estimate and radii and the second estimates the weights, fw j g m j=1 , of the linear regression model f(x) = m X j=1
Gas Plume Species Identification in LWIR Hyperspectral Imagery by Regression Analyses
Salvaggio, Carl
of the algorithm is a stepwise linear regression technique that only includes a basis vector in the model such as atmospheric compensation, gas absorption and emission, background modeling, and fitting a linear regression to a non-linear radiance model were addressed in order to generate the matrix of basis vectors. Synthetic
A regression model with a hidden logistic process for feature extraction from time series
Chamroukhi, Faicel
operation. The switch operations signals can be seen as time series presenting non-linearities and various changes in regime. Basic linear regression can not be adopted for this type of sig- nals because a constant linear relationship is not adapted. As alternative to linear regression, some authors use
Efficient Learning of Generalized Linear and Single Index Models with Isotonic Regression
Efficient Learning of Generalized Linear and Single Index Models with Isotonic Regression Sham M) provide powerful generalizations of linear regression, where the target variable is assumed to be a (possibly unknown) 1-dimensional function of a linear predictor. In gen- eral, these problems entail non
Sufficient reductions in regressions with elliptically contoured1 inverse predictors2
Bura, Efstathia
for21 the regression of Y on X comprises of a linear and a non-linear component.22 1 Introduction23 There are two general approaches based on inverse regression for estimating the linear sufficient9 reductions with18 parameters (µY , ) and density gY , there is no linear non-trivial sufficient reduction except
Bayesian Multivariate Poisson Regression for Models of Injury Count, by Severity
Kockelman, Kara M.
Bayesian Multivariate Poisson Regression for Models of Injury Count, by Severity By Jianming Ma, and lead to potential biases in sample databases. This paper offers a multivariate Poisson specification severity, Gibbs sampler, Markov chain Monte Carlo (MCMC) simulation, multivariate Poisson regression #12
Practical High Breakdown Regression David J. Olive and Douglas M. Hawkins
Olive, David
Practical High Breakdown Regression David J. Olive and Douglas M. Hawkins Southern Illinois breakdown n consistent regression es- timators exist. The response plot of the fitted values versus@umn.edu), School of Statistics, University of Minnesota, Minneapolis, MN 55455-0493, USA. Their work was supported
Kissock, Kelly
% of the world's energy consumption (Boyle 2004). The use of fossil fuels is the primary contributor to global, and production data to understand a company's energy performance over time. The method uses regression modelingUnderstanding Industrial Energy Use through Sliding Regression Analysis Carl W. Eger III, City
Regression-based estimation of ERP waveforms: I. The rERP framework
Kutas, Marta
Regression-based estimation of ERP waveforms: I. The rERP framework NATHANIEL J. SMITHa AND MARTA Science and Neurosciences, University of California, San Diego, San Diego, California, USA Abstract ERP of experimental designs. We introduce the regression-based rERP framework, which extends ERP averaging to handle
Exploiting Covariate Similarity in Sparse Regression via the Pairwise Elastic Net
Low, Steven H.
. Furthermore, un- like the Lasso, the Elastic Net can yield a sparse esti- mate with more than n non-zero477 Exploiting Covariate Similarity in Sparse Regression via the Pairwise Elastic Net Alexander to regression regulariza- tion called the Pairwise Elastic Net is pro- posed. Like the Elastic Net, it simultane
Ahn, Hongshik
-dimensional data using the logistic regression model as a base clas- sifier. CERP is similar to random subspace (Ho classification problems using ensem- bles of multinomial logistic regression models. A multinomial logit model is used as a base classifier in ensembles from random partitions of predictors. The multinomial logit
Non-parametric regression and neural-network inll drilling recovery models for carbonate reservoirs
Valkó, Peter
Non-parametric regression and neural-network in®ll drilling recovery models for carbonate This work introduces non-parametric regression and neural network models for forecasting the in®ll drilling variables. This situation mandates proper selection of independent variables for the in®ll drilling recovery
Ungar, Lyle H.
boolean and real-valued, are generated by structured search in the space of queries to the database of features automatically generated from a relational database. Structural Logistic Regression is an "upgradeIn Multi-Relational Data Mining Workshop at KDD-2003. Structural Logistic Regression for Link
Rothermel, Gregg
Regression testing is an important activity that can account for a large proportion of the cost of software testing technique that (1) chooses a subset of a test suite that was used to test the software before the modifications, and then (2) uses this subset to test the modified software. Selective regression testing
Distribution to Distribution Regression Junier B. Oliva joliva@cs.cmu.edu
Schneider, Jeff
Distribution to Distribution Regression Junier B. Oliva joliva@cs.cmu.edu Barnab´as P `Distribution to Distribution re- gression' where one is regressing a mapping where both the covariate (inputs) and re- sponse (outputs) are distributions. No pa- rameters on the input or output distributions
Regional regression models of watershed suspended-sediment discharge for the eastern United States
Vogel, Richard M.
: Sediment transport Regression Water quality Ungaged GAGES SPARROW s u m m a r y Estimates of mean annual Streamflow (GAGES) database. The resulting regional regression models summarized for major US water resources contaminants including pesticides, met- als, and polycyclic aromatic hydrocarbons (PAHs) readily sorb
Collins, J.L.
2004-12-02T23:59:59.000Z
The main objective of the Depleted UO{sub 2} Kernels Production Task at Oak Ridge National Laboratory (ORNL) was to conduct two small-scale production campaigns to produce 2 kg of UO{sub 2} kernels with diameters of 500 {+-} 20 {micro}m and 3.5 kg of UO{sub 2} kernels with diameters of 350 {+-} 10 {micro}m for the U.S. Department of Energy Advanced Fuel Cycle Initiative Program. The final acceptance requirements for the UO{sub 2} kernels are provided in the first section of this report. The kernels were prepared for use by the ORNL Metals and Ceramics Division in a development study to perfect the triisotropic (TRISO) coating process. It was important that the kernels be strong and near theoretical density, with excellent sphericity, minimal surface roughness, and no cracking. This report gives a detailed description of the production efforts and results as well as an in-depth description of the internal gelation process and its chemistry. It describes the laboratory-scale gel-forming apparatus, optimum broth formulation and operating conditions, preparation of the acid-deficient uranyl nitrate stock solution, the system used to provide uniform broth droplet formation and control, and the process of calcining and sintering UO{sub 3} {center_dot} 2H{sub 2}O microspheres to form dense UO{sub 2} kernels. The report also describes improvements and best past practices for uranium kernel formation via the internal gelation process, which utilizes hexamethylenetetramine and urea. Improvements were made in broth formulation and broth droplet formation and control that made it possible in many of the runs in the campaign to produce the desired 350 {+-} 10-{micro}m-diameter kernels, and to obtain very high yields.
Reality - an emerging representation of the world
Martin A. Green
2009-03-11T23:59:59.000Z
Some unique source -- the world, W -- must underlie all the information realized in the universe throughout time. Perceived reality, R, is a progressively emerging representation of W in the form of the geometrical universe. Time corresponds to the process of emergence. When first represented in R, information about W is expressed in a non-localized, quantum manner. As the emergence proceeds, most information becomes inaccessible (entropy), supporting the robust, redundant encoding of accessible records. The past is encoded in and inferred from present records; the anticipated future will preserve present information and reveal unpredictable new information about W. Emergence of the future demands non-unitary reduction of quantum states and increased Kolmogorov complexity of the quasi-classical records in terms of which the quantum states are known. Given the limited information content of records, the quasi-classical universe lacks fine details; whereas the future must be uncertain to admit new information.
A representation formula for maps on supermanifolds
Helein, Frederic [Institut de Mathematiques de Jussieu, UMR 7586, Universite Denis Diderot-Paris 7, Case 7012, 2 place Jussieu, 75251 Paris Cedex 5 (France)
2008-02-15T23:59:59.000Z
We analyze the notion of morphisms of rings of superfunctions which is the basic concept underlying the definition of supermanifolds as ringed spaces (i.e., following Berezin, Leites, Manin, etc.). We establish a representation formula for all (pull-back) morphisms from the algebra of functions on an ordinary manifolds to the superalgebra of functions on an open subset of a superspace. We then derive two consequences of this result. The first one is that we can integrate the data associated with a morphism in order to get a (nonunique) map defined on an ordinary space (and uniqueness can be achieved by restriction to a scheme). The second one is a simple and intuitive recipe to compute pull-back images of a function on a manifold M by a map from a superspace to M.
PURE MATHEMATICS Representation theory(Baranov, Pirashvili, Snashall)
and theoretical physics. Much of the development of modern representation theory was motivated by representations'. A surprising amount of classical ge- ometric techniques and concepts (such as smoothness, homo- geneous spaces mathematical notion has its roots in mathematical physics, in particular exactly soluble models of statistical
2004-09-16 | Greg Jane 1 Semantic representation
Janée, Greg
2004-09-16 | Greg Janée 1 Semantic representation · Semantic definitions explicit representations property rights programmatic interpretation would be beneficial · Here's a common framework for both #12;2004-09-16 | Greg Janée 2 Archival item facets · (Identifier) · Metadata · Data · Intellectual property rights #12;2004
Document Representation and Query Expansion Models for Blog Recommendation
Callan, Jamie
Document Representation and Query Expansion Models for Blog Recommendation Jaime Arguello document representation models and two query expansion models for the task of recommend- ing blogs to a user in response to a query. Blog relevance ranking differs from traditional document ranking in ad
Deep Learning Representation using Autoencoder for 3D Shape Retrieval
benchmarks. I. INTRODUCTION With the fast development of 3D printer, Microsoft Kinect sensor and laserDeep Learning Representation using Autoencoder for 3D Shape Retrieval Zhuotun Zhu, Xinggang Wang@hust.edu.cn Abstract--We study the problem of how to build a deep learning representation for 3D shape. Deep learning
Closing the Learning-Planning Loop with Predictive State Representations
Guestrin, Carlos
Closing the Learning-Planning Loop with Predictive State Representations Byron Boots Machine and sta- tistically consistent spectral algorithm for learning the pa- rameters of a Predictive State the essential features of the environment. This representation allows accurate prediction with a small number
SEMANTIC DATA INTEGRATION IN A MULTIPLE REPRESENTATION ENVIRONMENT
KĂ¶bben, Barend
SEMANTIC DATA INTEGRATION IN A MULTIPLE REPRESENTATION ENVIRONMENT J.E. Stotera , R.L.G. Lemmensa: Semantic data integration, Multi Representation Database, Generalisation, ontologies, machine ontology at the different scales are semantically integrated, 2) objects in the different scales representing the same real
Bringing Together Human and Robotic Environment Representations A Pilot Study
Bringing Together Human and Robotic Environment Representations A Pilot Study Elin A. Topp, Helge and Communication (CSC) Royal Institute of Technology (KTH) 10044 Stockholm, Sweden Email: {topp,hehu,hic,kse}@csc.kth.se Abstract-- Human interaction with a service robot requires a shared representation of the environment
Fast Multipole Representation of Diffusion Curves and Points Timothy Sun
Grinspun, Eitan
Fast Multipole Representation of Diffusion Curves and Points Timothy Sun Papoj Thamjaroenporn performed on the fast multipole representation. Abstract We propose a new algorithm for random-access evaluation of diffu- sion curve images (DCIs) using the fast multipole method. Unlike all previous methods
Using Graphical Representations to Support the Calculation of Infusion Parameters
Subramanian, Sriram
Using Graphical Representations to Support the Calculation of Infusion Parameters Sandy J. J. Gould in which participants were asked to solve a num- ber of infusion parameter problems that were represented representations transfer to actual workplace settings. Keywords: Graphical reasoning, infusion pumps, re
Using Compact Data Representations for Languages Based on Catamorphisms
Fegaras, Leonidas
Using Compact Data Representations for Languages Based on Catamorphisms OGI, Techreport #95 on using a compact vector representation for the recursive structure over which the catamorphism operates program performance, in particular by performing program fusion [13, 3, 7, 1]. In this paper, we explore
Systematic Approach to the Design of RepresentationChanging Algorithms
Fink, Eugene
The performance of all problemsolving sys tems depends crucially on problem representa tion. The same problem representation in an AI problemsolving system is the input to the system. In most problem solving systems problemsolving systems de pends crucially on problem representation. The same problem may be easy or di
Regression Models for Demand Reduction based on Cluster Analysis of Load Profiles
Yamaguchi, Nobuyuki; Han, Junqiao; Ghatikar, Girish; Piette, Mary Ann; Asano, Hiroshi; Kiliccote, Sila
2009-06-28T23:59:59.000Z
This paper provides new regression models for demand reduction of Demand Response programs for the purpose of ex ante evaluation of the programs and screening for recruiting customer enrollment into the programs. The proposed regression models employ load sensitivity to outside air temperature and representative load pattern derived from cluster analysis of customer baseline load as explanatory variables. The proposed models examined their performances from the viewpoint of validity of explanatory variables and fitness of regressions, using actual load profile data of Pacific Gas and Electric Company's commercial and industrial customers who participated in the 2008 Critical Peak Pricing program including Manual and Automated Demand Response.
C.P.Oertel; J.R.Giles
2009-11-01T23:59:59.000Z
Characterization of radionuclide concentrations in soil profiles requires accurate evaluation of the depth distribution of the concentrations as measured by gamma emissions. An ongoing study based on 137Cs activity has shown that such concentration data generally follow an exponential trend when the fraction of radioactivity below depth is plotted against the depth. The slope of the exponential regression fit is defined as alpha/rho, the depth profile parameter. A weighted exponential regression procedure has been developed to compute a mean ??? for a group of related soil samples. Regression results from different areas or from different time periods can be used to compare representative radionuclide concentrations for the specified groupings.
On the representation of many-body interactions in water
Medders, Gregory R; Morales, Miguel A; Paesani, Francesco
2015-01-01T23:59:59.000Z
Recent work has shown that the many-body expansion of the interaction energy can effectively be used to develop analytical representations of global potential energy surfaces (PESs) for water. In this study, the role of short- and long-range contri- butions at different orders is investigated by analyzing water potentials that treat the leading terms of the many-body expansion through implicit (i.e., TTM3-F and TTM4-F PESs) and explicit (i.e., WHBB and MB-pol PESs) representations. It is found that explicit short-range representations of 2-body and 3-body interactions along with a physically correct integration of short- and long-range contributions are necessary for an accurate representation of the water interactions from the gas to the condensed phase. Similarly, a complete many-body representation of the dipole moment surface is found to be crucial to reproducing the correct intensities of the infrared spectrum of liquid water.
Katipamula, S.; Reddy, T. A.; Claridge, D. E.
1994-01-01T23:59:59.000Z
An empirical or regression modeling approach is simple to develop and easy to use compared to use of detailed hourly simulations. Therefore, regression analysis has become a widely used tool in the determination of annual energy savings accruing...
Operator algebras and conformal eld theory III. Fusion of positive energy representations
Proudfoot, Nicholas
Operator algebras and conformal ®eld theory III. Fusion of positive energy representations of LSU ........................................................................ 513 V. Connes fusion of positive energy representations. (4) A computable and manifestly unitary de®nition of fusion for positive energy representations
Predicting Turbulence using Partial Least Squares Regression and an Artificial Neural Network
Lakshmanan, Valliappa
Predicting Turbulence using Partial Least Squares Regression and an Artificial Neural Network in the dataset. Then, the transformed data are pre- sented to a neural network whose output node has a sigmoid
Reddy, T. A.; Claridge, D.; Wu, J.
1992-01-01T23:59:59.000Z
Statistical models of energy use in commercial buildings are being increasingly used not only for predicting retrofit savings but also for identifying improper operation of HVAC systems. The conventional approach involves using multiple regression...
Modeling Personalized Email Prioritization: Classification-based and Regression-based Approaches
Yoo S.; Yang, Y.; Carbonell, J.
2011-10-24T23:59:59.000Z
Email overload, even after spam filtering, presents a serious productivity challenge for busy professionals and executives. One solution is automated prioritization of incoming emails to ensure the most important are read and processed quickly, while others are processed later as/if time permits in declining priority levels. This paper presents a study of machine learning approaches to email prioritization into discrete levels, comparing ordinal regression versus classier cascades. Given the ordinal nature of discrete email priority levels, SVM ordinal regression would be expected to perform well, but surprisingly a cascade of SVM classifiers significantly outperforms ordinal regression for email prioritization. In contrast, SVM regression performs well -- better than classifiers -- on selected UCI data sets. This unexpected performance inversion is analyzed and results are presented, providing core functionality for email prioritization systems.
Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression
Chen, Wan-Ping
2015-01-01T23:59:59.000Z
in high-dimensional multivariate regression. ” The Annals ofSupport Union Recovery in Multivariate Multi-Response LinearSupport Union Recovery in Multivariate Multi-Response Linear
Greenberg, Albert
Iterative Multivariate Regression Model for Correlated Responses Prediction S. Tom Au, Guangqin Ma- tive procedure to model multiple responses prediction into correlated multivariate predicting scheme, which is always favorable for responses separations in our multivariate prediction. We also point out
Frank, M. D.; Beattie, B. R.
1979-01-01T23:59:59.000Z
inventory, value of livestock inventory and miscellaneous expenditures. Using 1969 Census of Agriculture data, each regional function was statistically fit using both ordinary least squares (OLS) and ridge regression. As expected, parameter estimates under...