Saved in:
Bibliographic Details
Main Authors: Zhang, Lujing, Hsu, Daniel, Balakrishnan, Sivaraman
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.07319
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914378425565184
author Zhang, Lujing
Hsu, Daniel
Balakrishnan, Sivaraman
author_facet Zhang, Lujing
Hsu, Daniel
Balakrishnan, Sivaraman
contents Multi-group learning is a learning task that focuses on controlling predictors' conditional losses over specified subgroups. We propose ShakyPrepend, a method that leverages tools inspired by differential privacy to obtain improved theoretical guarantees over existing approaches. Through numerical experiments, we demonstrate that ShakyPrepend adapts to both group structure and spatial heterogeneity. We provide practical guidance for deploying multi-group learning algorithms in real-world settings.
format Preprint
id arxiv_https___arxiv_org_abs_2603_07319
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ShakyPrepend: A Multi-Group Learner with Improved Sample Complexity
Zhang, Lujing
Hsu, Daniel
Balakrishnan, Sivaraman
Machine Learning
Multi-group learning is a learning task that focuses on controlling predictors' conditional losses over specified subgroups. We propose ShakyPrepend, a method that leverages tools inspired by differential privacy to obtain improved theoretical guarantees over existing approaches. Through numerical experiments, we demonstrate that ShakyPrepend adapts to both group structure and spatial heterogeneity. We provide practical guidance for deploying multi-group learning algorithms in real-world settings.
title ShakyPrepend: A Multi-Group Learner with Improved Sample Complexity
topic Machine Learning
url https://arxiv.org/abs/2603.07319