Choi, Seungjin - Department of Computer Science and Engineering, Pohang University of Science and Technology

• Orthogonal Nonnegative Matrix Factorization: Multiplicative Updates on Stiefel Manifolds
• Differential Hebbiantype Learning Algorithms for Decorrelation and Independent Component Analysis
• On Spectral Basis Selection for Single Channel Polyphonic Music Separation
• A HYBRID LEARNING APPROACH TO BLIND DECONVOLUTION OF MIMO Seungjin CHOI y and Andrzej CICHOCKI z
• APPLICATION OF NON-NEGATIVE MATRIX FACTORIZATION TO DYNAMIC POSITRON
• MULTIPLE NONNEGATIVE-MATRIX FACTORIZATION OF DYNAMIC PET IMAGES Jong-Hoon Ahn
• SEQUENTIAL EM LEARNING FOR SUBSPACE ANALYSIS Seungjin Choi
• Kernel Isomap Heeyoul Choi and Seungjin Choi 1
• CONSTRAINED MINIMUM ENTROPY AND MAXIMUM NEGENTROPY BLIND DECONVOLUTION AND EQUALIZATION
• TARGET SPEECH EXTRACTION WITH LEARNED SPECTRAL BASES Sunho Park, Jiho Yoo, and Seungjin Choi
• Algorithms for Orthogonal Nonnegative Matrix Factorization Seungjin Choi
• Robust kernel Isomap Heeyoul Choi , Seungjin Choi
• BLIND SEPARATION OF SECONDORDER NONSTATIONARY AND TEMPORALLY COLORED SOURCES
• Constrained Projection Approximation Algorithms for Principal Component Analysis
• Manifold Integration with Markov Random Walks Heeyoul Choi
• IEEE SIGNAL PROCESSING LETTERS, VOL. 14, NO. 10, OCTOBER 2007 1 Two-Dimensional Canonical Correlation Analysis
• F-SEONS: A Second-Order Frequency-Domain Algorithm for Noisy Convolutive Source Separation
• BLIND SEPARATION OF DELAYED AND SUPERIMPOSED ACOUSTIC SOURCES: LEARNING ALGORITHM AND EXPERIMENTAL STUDY
• Third IEEE Signal Processing Workshop on Signal Processing Advances in Wireless Communications,Taoyuan, Taiwan, March 20-23, 2001 APPROXIMATE MAXIMUM LIKELIHOOD SOURCE SEPARATION USING
• RECTIFIED SUBSPACE ANALYSIS OF DYNAMIC POSITRON EMISSION TOMOGRAPHY Seungjin Choi x , Sangki Kim x , Jae Sung Lee y , Dong Soo Lee y
• Comparative evaluation of CASA and BSS models for subband cocktailparty speech separation
• A CASA FRONT-END USING THE LOCALISATION CUE FOR SEGREGATION AND THEN COCKTAIL-PARTY SPEECH RECOGNITION
• Contrast Enhancement of Missile Video Sequences via Image Stabilization and Product Correlation
• Topographic Independent Component Analysis of Gene Expression Time Series Data
• Sequential EM Learning for Subspace Seungjin Choi
• ACOUSTIC SOURCE SEPARATION: FUNDAMENTAL ISSUES Seungjin CHOI
• DIFFERENTIAL DECORRELATION FOR NONSTATIONARY SOURCE SEPARATION Seungjin Choi x , Andrzej Cichocki y , Yannick Deville z
• IEICE TRANS. FUNDAMENTALS, VOL.E85 {A, NO.12 DECEMBER 2002 Approximate Maximum Likelihood Source Separation
• Journal of VLSI Signal Processing, ?, 1--13 (2001) 2001 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
• A SYMMETRICDEFINITE PENCIL APPROACH TO SOURCE SEPARATION Seungjin Choi
• A METHOD OF INITIALIZATION FOR NONNEGATIVE MATRIX FACTORIZATION Yong-Deok Kim and Seungjin Choi
• INDEPENDENT SUBSPACES OF GENE EXPRESSION DATA Intelligent Robot Research Division
• Multichannel Signal Separation for Cocktail Party Speech Recognition: A Dynamic
• BLIND SEPARATION OF NONSTATIONARY AND TEMPORALLY CORRELATED SOURCES
• NONSTATIONARY SOURCE SEPARATION Seungjin CHOI and O Young LEE
• Color Face Tensor Factorization and Slicing for Illumination-Robust Recognition
• IEICE TRANS. FUNDAMENTALS, VOL.E83 {A, NO.12 DECEMBER 2000 Natural Gradient Learning for Spatio-temporal
• PCABASED LINEAR DYNAMICAL SYSTEMS FOR MULTICHANNEL EEG CLASSIFICATION
• Blind Separation of Nonstationary Sources in Noisy Mixtures Seungjin CHOI x1 and Andrzej CICHOCKI y
• Nonnegative Matrix Factorization for Motor Imagery EEG Classification
• Kernel Design II Department of Computer Science
• WHY BLIND SOURCE SEPARATION FOR BLIND EQUALIZATION OF MULTIPLE Inseon Jang and Seungjin Choi
• Adaptive Di erential Decorrelation: A Natural Gradient Algorithm
• Equivariant Nonstationary Source Separation Seungjin Choi, x1 Andrzej Cichocki, y and Shunichi Amari y
• Fast Stochastic Neighbor Embedding: A Trust-Region Algorithm
• A Hybrid Learning Approach to Blind Deconvolution of Linear MIMO Systems
• NON-NEGATIVE COMPONENT PARTS OF SOUND FOR CLASSIFICATION Yong-Choon Cho, Seungjin Choi, Sung-Yang Bang
• Trust-Region Learning for ICA Heeyoul Choi, Sookjeong Kim, Seungjin Choi
• Trust-region methods on Riemannian manifolds
• Cascade Neural Networks for Multichannel Blind Deconvolution Seungjin CHOI \Lambda 1 and Andrzej CICHOCKI \Lambda\Lambda
• Independent Arrays or Independent Time Courses for Gene Expression Time Series
• Simpler Knowledge-based Support Vector Machines
• A Relative Trust-Region Algorithm for Independent Component Analysis
• NONNEGATIVE TUCKER DECOMPOSITION WITH ALPHA-DIVERGENCE Yong-Deok Kim
• TO APPEAR IN JOURNAL OF COMMUNICATIONS AND NETWORKS, 1999 1 An Unsupervised Hybrid Network for Blind
• CUR+NMF for Learning Spectral Features from Large Data Matrix Hyekyoung Lee and Seungjin Choi
• Learning Principal Directions: Integrated-Squared-Error Minimization
• Wake-Sleep PCA Seungjin Choi, Member, IEEE
• Kernel Design Department of Computer Science
• Regularization Department of Computer Science
• MULTICHANNEL SIGNAL SEPARATION FOR COCKTAIL PARTY SPEECH RECOGNITION: A DYNAMIC RECURRENT NETWORK
• PROBABILISTIC MATRIX TRI-FACTORIZATION Jiho Yoo and Seungjin Choi
• A Spurious Equilibriafree Learning Algorithm for the Blind Separation of Nonzero Skewness Signals
• BLIND SOURCE SEPARATION FROM NOISY MEASUREMENTS: TIME-DELAYED CORRELATION MATCHING
• Relative Gradient Learning for Independent Subspace Analysis Heeyoul Choi and Seungjin Choi
• A Coupled Helmholtz Machine for PCA Seungjin Choi 1
• Sequence-driven features for prediction of subcellular localization of proteins
• Kernel Isomap on Noisy Manifold Heeyoul Choi, Seungjin Choi
• Blind Source Separation and Independent Component Analysis: Seungjin Choi
• CORRELATION MATCHING APPROACH TO SOURCE SEPARATION IN THE PRESENCE OF SPATIALLY CORRELATED NOISE
• Evaluation of CASA and BSS models for subband cocktailparty speech separation
• Incremental Learning to Rank with Partially-Labeled Data Kye-Hyeon Kim
• A Clinical Application of Ensemble ICA to the Quantification of Myocardial Blood Flow in Dynamic
• U-Likelihood and U-Updating Algorithms: Statistical Inference in Latent Variable Models
• Journal of the Korean Physical Society, Vol. 49, No. 4, October 2006, pp. 16671674 Development of a Separation Algorithm for Peak Signals and Its Application
• TWO SPATIOTEMPORAL DECORRELATION LEARNING ALGORITHMS AND THEIR APPLICATION TO MULTICHANNEL BLIND DECONVOLUTION
• Algebraic Di erential Decorrelation for Nonstationary Source Separation
• A Multiplicative Up-Propagation Algorithm Jong-Hoon Ahn jonghun@postech.ac.kr
• A COMPARISION OF PCA/ICA FOR DATA PREPROCESSING IN A GEOSCIENCE APPLICATION
• A Bayesian Network Classifier and Hierarchical Gabor Features for Handwritten
• SELFADAPTIVE INDEPENDENT COMPONENT ANALYSIS for SUBGAUSSIAN and SUPERGAUSSIAN MIXTURES with
• LOCAL STABILITY ANALYSIS OF FLEXIBLE INDEPENDENT COMPONENT ANALYSIS ALGORITHM
• NATURAL GRADIENT LEARNING WITH A NONHOLONOMIC CONSTRAINT FOR BLIND DECONVOLUTION OF MULTIPLE CHANNELS
• Blind Equalization via Approximate Maximum Likelihood Source Seungjin CHOI x1 and Andrzej CICHOCKI y
• TO APPEAR IN IEEE SIGNAL PROCESSING LETTERS, FEBRUARY 10, 2007 1 Two-Dimensional Canonical Correlation Analysis
• Differential Learning Algorithms for Decorrelation and Independent Component Analysis
• Semidefinite spectral clustering Jaehwan Kim a
• Maximum Within-Cluster Association Yongjin Lee, Seungjin Choi
• Nonnegative Features of Spetro-Temporal Sounds for Classification
• ENSEMBLES OF LANDMARK MULTIDIMENSIONAL SCALINGS Seunghak Lee
• GAUSSIAN PROCESSES FOR SOURCE SEPARATION Sunho Park and Seungjin Choi
• Source Separation with Gaussian Process Models Sunho Park and Seungjin Choi
• Nonnegative Tucker Decomposition Yong-Deok Kim, Seungjin Choi
• NON-NEGATIVE TENSOR FACTORIZATION USING ALPHA AND BETA DIVERGENCES Andrzej CICHOCKI1
• SOFT GEODESIC KERNEL K-MEANS Jaehwan Kim
• Novel Multi-layer Non-negative Tensor Factorization with Sparsity Constraints
• Tree-Dependent Components of Gene Expression Data for Clustering
• ICA-Based Clustering for Resolving Permutation Ambiguity in Frequency-Domain Convolutive Source Separation
• Rao-Blackwellized Particle Filtering for Sequential Speech Enhancement
• ITERATIVE PROJECTION APPROXIMATION ALGORITHMS FOR PCA Seungjin Choi
• Monaural Music Source Separation: Nonnegativity, Sparseness, and Shift-Invariance
• On Variations of Power Iteration Seungjin Choi
• Myocardial Blood Flow Quantification in Dynamic PET: An Ensemble ICA Approach
• RELATIVE TRUST-REGION LEARNING FOR ICA Heeyoul Choi, Seungjin Choi
• Learning Nonnegative Features of Spectro-Temporal Sounds for Classification Yong-Choon Cho
• Minimum Entropy, k-Means, Spectral Clustering Yongjin Lee
• A NEW WAY OF PCA: INTEGRATED-SQUARED-ERROR AND EM ALGORITHMS Jong-Hoon Ahn
• ICA+OPCA FOR ARTIFACT-ROBUST CLASSIFICATION OF EEG DATA
• Membership Scoring via Independent Feature Subspace Analysis for Grouping Co-Expressed Genes
• PCA+HMM+SVM FOR EEG PATTERN CLASSIFICATION Hyekyung Lee and Seungjin Choi
• Differential ICA Seungjin Choi
• DIFFERENTIAL LEARNING AND RANDOM WALK MODEL Seungjin Choi
• Several improvements of the Hrault-Jutten model for speech segregation
• SEQUENTIAL LEAST SQUARES ALGORITHMS FOR BLIND CO-CHANNEL SIGNAL Inseon Jang and Seungjin Choi
• Abstract--Recently suggested non-negative matrix factorization (NMF) seems to overcome fundamental limitations
• Abstract--For the quantitation of cerebral blood flow (CBF) O PET, the measurement of arterial input function
• MachineLearningGroup DepartmentofComputerScience,POSTECH
• Prediction of subcellular localization of proteins using pairwise sequence alignment
• WEIGHTED NONNEGATIVE MATRIX FACTORIZATION Yong-Deok Kim and Seungjin Choi
• Color Face Tensor Factorization and Slicing for Illumination-Robust Recognition
• Nonnegative Tucker Decomposition Yong-Deok Kim, Seungjin Choi
• A METHOD OF INITIALIZATION FOR NONNEGATIVE MATRIX FACTORIZATION Yong-Deok Kim and Seungjin Choi
• Nonnegative matrix factorization with -divergence
• International Journal of Neural Systems Special Issue on Issue's Topic
• Gaussian Process Regression for Voice Activity Detection and Speech Enhancement
• Neighbor Search with Global Geometry: A Minimax Message Passing Algorithm
• Dynamically Weighted Hidden Markov Model for Spam Deobfuscation Seunghak Lee
• Matrix Co-Factorization on Compressed Sensing and Seungjin Choi1,2
• Restricted Deep Belief Networks for Multi-View Learning
• MachineLearningGroup DepartmentofComputerScience,POSTECH
• MachineLearningGroup DepartmentofComputerScience,POSTECH
• 2011 IEEE International Workshop on Machine Learning for Signal Processing September 18-21, 2011, Beijing, China
• Nystrom Approximations for Scalable Face Recognition: A Comparative Study
• GEOMETRIC PROGRAMMING FOR AGGREGATION OF BINARY CLASSIFIERS Sunho Park 1
• MULTIPLE KERNEL NONNEGATIVE MATRIX FACTORIZATION Shounan An 1
• Kang Hyohyeong PhD Student,
• Bayesian Multi-Task Learning for Common Spatial Patterns Hyohyeong Kang