Feature-distributed sparse regression: a screen-and-clean approach
Jiyan Yang, Michael W. Mahoney, Michael Saunders, Yuekai Sun
Abstract
Most existing approaches to distributed sparse regression assume the data is partitioned by samples. However, for high-dimensional data (D >> N), it is more natural to partition the data by features. We propose an algorithm to distributed sparse regression when the data is partitioned by features rather than samples. Our approach allows the user to tailor our general method to various distributed computing platforms by trading-off the total amount of data (in bits) sent over the communication network and the number of rounds of communication. We show that an implementation of our approach is capable of solving L1-regularized L2 regression problems with millions of features in minutes.
BibTeX
@inproceedings{NIPS2016_363763e5,
author = {Yang, Jiyan and Mahoney, Michael W and Saunders, Michael and Sun, Yuekai},
booktitle = {Advances in Neural Information Processing Systems},
editor = {D. Lee and M. Sugiyama and U. Luxburg and I. Guyon and R. Garnett},
pages = {},
publisher = {Curran Associates, Inc.},
title = {Feature-distributed sparse regression: a screen-and-clean approach},
url = {https://proceedings.neurips.cc/paper_files/paper/2016/file/363763e5c3dc3a68b399058c34aecf2c-Paper.pdf},
volume = {29},
year = {2016}
}