skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Iterative random forests to discover predictive and stable high-order interactions

Authors:
; ; ;
Publication Date:
Sponsoring Org.:
USDOE
OSTI Identifier:
1417528
Grant/Contract Number:
DOE DE-AC02-05CH11231; SC0017069
Resource Type:
Journal Article: Published Article
Journal Name:
Proceedings of the National Academy of Sciences of the United States of America
Additional Journal Information:
Related Information: CHORUS Timestamp: 2018-01-19 14:11:14; Journal ID: ISSN 0027-8424
Publisher:
Proceedings of the National Academy of Sciences
Country of Publication:
United States
Language:
English

Citation Formats

Basu, Sumanta, Kumbier, Karl, Brown, James B., and Yu, Bin. Iterative random forests to discover predictive and stable high-order interactions. United States: N. p., 2018. Web. doi:10.1073/pnas.1711236115.
Basu, Sumanta, Kumbier, Karl, Brown, James B., & Yu, Bin. Iterative random forests to discover predictive and stable high-order interactions. United States. doi:10.1073/pnas.1711236115.
Basu, Sumanta, Kumbier, Karl, Brown, James B., and Yu, Bin. 2018. "Iterative random forests to discover predictive and stable high-order interactions". United States. doi:10.1073/pnas.1711236115.
@article{osti_1417528,
title = {Iterative random forests to discover predictive and stable high-order interactions},
author = {Basu, Sumanta and Kumbier, Karl and Brown, James B. and Yu, Bin},
abstractNote = {},
doi = {10.1073/pnas.1711236115},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
number = ,
volume = ,
place = {United States},
year = 2018,
month = 1
}

Journal Article:
Free Publicly Available Full Text
This content will become publicly available on July 19, 2018
Publisher's Version of Record

Save / Share:
  • An iterative implicit-explicit hybrid scheme is proposed for one-dimensional ideal magnetohydrodynamical (MHD) equations. The scheme is accurate to second order in both space and time for all Courant numbers. Each kind of MHD wave, both compressible and incompressible, may be either implicitly or explicitly, or partially implicitly and partially explicitly treated depending on their associated Courant numbers in each numerical cell, and the scheme is able to smoothly switch between explicit and implicit calculations. The scheme is of Godunov type in both explicit and implicit regimes, in which the flux needed in a Godunov scheme is calculated from Riemann problems,more » is in a strictly conservation form, and is able to resolve MHD discontinuities. Only a single level of iterations is involved in the scheme, and the iterations solve both the implicit relations arising from upstream centered differences for all wave families and the nonlinearity of MHD equations. Multicolors proposed in the paper may significantly reduce the number of iterations required to reach a converged solution. Compared with the largest Courant number in a simulation, only a small number of iterations are needed in the scheme. The feature of the scheme has been shown through numerical examples. It is easy to vectorize the computer code of the scheme.« less
  • The high-order low-order (HOLO) method is a recently developed moment-based acceleration scheme for solving time-dependent thermal radiative transfer problems, and has been shown to exhibit orders of magnitude speedups over traditional time-stepping schemes. However, a linear stability analysis by Haut et al. (2015 Haut, T. S., Lowrie, R. B., Park, H., Rauenzahn, R. M., Wollaber, A. B. (2015). A linear stability analysis of the multigroup High-Order Low-Order (HOLO) method. In Proceedings of the Joint International Conference on Mathematics and Computation (M&C), Supercomputing in Nuclear Applications (SNA) and the Monte Carlo (MC) Method; Nashville, TN, April 19–23, 2015. American Nuclear Society.)more » revealed that the current formulation of the multigroup HOLO method was unstable in certain parameter regions. Since then, we have replaced the intensity-weighted opacity in the first angular moment equation of the low-order (LO) system with the Rosseland opacity. Furthermore, this results in a modified HOLO method (HOLO-R) that is significantly more stable.« less
  • Habitat prediction models for rare plants can be useful when large areas must be surveyed or populations must be established. Investigators developed a habitat prediction model for four species of Southeastern hardwood forests. These four examples suggest that models based on resource and vegetation characteristics can accurately predict habitat, but only when plants are strongly associated with these variables and the scale of modeling coincides with habitat size.
  • High-field magnetization curves of the series of amorphous alloys Dy{sub {ital x}}Gd{sub 1{minus}{ital x}}Ni have been measured up to 15 tesla and interpreted in the framework of the theory of Chudnovsky {ital et} {ital al}. The results show several features (high-field crossovers, correlation volume of local-anisotropy directions, ferromagnetic correlation length) all consistent with each other and in agreement with theoretical predictions.