Back to Search Start Over

Fast Simultaneous Training of Generalized Linear Models (FaSTGLZ)

Authors :
Conroy, Bryan R.
Walz, Jennifer M.
Cheung, Brian
Sajda, Paul
Publication Year :
2013

Abstract

We present an efficient algorithm for simultaneously training sparse generalized linear models across many related problems, which may arise from bootstrapping, cross-validation and nonparametric permutation testing. Our approach leverages the redundancies across problems to obtain significant computational improvements relative to solving the problems sequentially by a conventional algorithm. We demonstrate our fast simultaneous training of generalized linear models (FaSTGLZ) algorithm on a number of real-world datasets, and we run otherwise computationally intensive bootstrapping and permutation test analyses that are typically necessary for obtaining statistically rigorous classification results and meaningful interpretation. Code is freely available at http://liinc.bme.columbia.edu/fastglz.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.1307.8430
Document Type :
Working Paper