A Polynomial-Time Form of Robust Regression
Abstract
Despite the variety of robust regression methods that have been developed, current regression formulations are either NP-hard, or allow unbounded response to even a single leverage point. We present a general formulation for robust regression --Variational M-estimation--that unifies a number of robust regression methods while allowing a tractable approximation strategy. We develop an estimator that requires only polynomial-time, while achieving certain robustness and consistency guarantees. An experimental evaluation demonstrates the effectiveness of the new estimation approach compared to standard methods.
Cite
Text
Yu et al. "A Polynomial-Time Form of Robust Regression." Neural Information Processing Systems, 2012.Markdown
[Yu et al. "A Polynomial-Time Form of Robust Regression." Neural Information Processing Systems, 2012.](https://mlanthology.org/neurips/2012/yu2012neurips-polynomialtime/)BibTeX
@inproceedings{yu2012neurips-polynomialtime,
title = {{A Polynomial-Time Form of Robust Regression}},
author = {Yu, Yao-liang and Aslan, Özlem and Schuurmans, Dale},
booktitle = {Neural Information Processing Systems},
year = {2012},
pages = {2483-2491},
url = {https://mlanthology.org/neurips/2012/yu2012neurips-polynomialtime/}
}