Saved in:
Bibliographic Details
Main Author: Holland, Matthew J.
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2301.11584
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917585239408640
author Holland, Matthew J.
author_facet Holland, Matthew J.
contents Under losses which are potentially heavy-tailed, we consider the task of minimizing sums of the loss mean and standard deviation, without trying to accurately estimate the variance. By modifying a technique for variance-free robust mean estimation to fit our problem setting, we derive a simple learning procedure which can be easily combined with standard gradient-based solvers to be used in traditional machine learning workflows. Empirically, we verify that our proposed approach, despite its simplicity, performs as well or better than even the best-performing candidates derived from alternative criteria such as CVaR or DRO risks on a variety of datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2301_11584
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Robust variance-regularized risk minimization with concomitant scaling
Holland, Matthew J.
Machine Learning
Under losses which are potentially heavy-tailed, we consider the task of minimizing sums of the loss mean and standard deviation, without trying to accurately estimate the variance. By modifying a technique for variance-free robust mean estimation to fit our problem setting, we derive a simple learning procedure which can be easily combined with standard gradient-based solvers to be used in traditional machine learning workflows. Empirically, we verify that our proposed approach, despite its simplicity, performs as well or better than even the best-performing candidates derived from alternative criteria such as CVaR or DRO risks on a variety of datasets.
title Robust variance-regularized risk minimization with concomitant scaling
topic Machine Learning
url https://arxiv.org/abs/2301.11584