Saved in:
Bibliographic Details
Main Authors: Hanneke, Steve, Moran, Shay, Shlimovich, Alexander, Yehudayoff, Amir
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.14572
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908338030116864
author Hanneke, Steve
Moran, Shay
Shlimovich, Alexander
Yehudayoff, Amir
author_facet Hanneke, Steve
Moran, Shay
Shlimovich, Alexander
Yehudayoff, Amir
contents Learning theory has traditionally followed a model-centric approach, focusing on designing optimal algorithms for a fixed natural learning task (e.g., linear classification or regression). In this paper, we adopt a complementary data-centric perspective, whereby we fix a natural learning rule and focus on optimizing the training data. Specifically, we study the following question: given a learning rule $\mathcal{A}$ and a data selection budget $n$, how well can $\mathcal{A}$ perform when trained on at most $n$ data points selected from a population of $N$ points? We investigate when it is possible to select $n \ll N$ points and achieve performance comparable to training on the entire population. We address this question across a variety of empirical risk minimizers. Our results include optimal data-selection bounds for mean estimation, linear classification, and linear regression. Additionally, we establish two general results: a taxonomy of error rates in binary classification and in stochastic convex optimization. Finally, we propose several open questions and directions for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2504_14572
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Data Selection for ERMs
Hanneke, Steve
Moran, Shay
Shlimovich, Alexander
Yehudayoff, Amir
Machine Learning
Learning theory has traditionally followed a model-centric approach, focusing on designing optimal algorithms for a fixed natural learning task (e.g., linear classification or regression). In this paper, we adopt a complementary data-centric perspective, whereby we fix a natural learning rule and focus on optimizing the training data. Specifically, we study the following question: given a learning rule $\mathcal{A}$ and a data selection budget $n$, how well can $\mathcal{A}$ perform when trained on at most $n$ data points selected from a population of $N$ points? We investigate when it is possible to select $n \ll N$ points and achieve performance comparable to training on the entire population. We address this question across a variety of empirical risk minimizers. Our results include optimal data-selection bounds for mean estimation, linear classification, and linear regression. Additionally, we establish two general results: a taxonomy of error rates in binary classification and in stochastic convex optimization. Finally, we propose several open questions and directions for future research.
title Data Selection for ERMs
topic Machine Learning
url https://arxiv.org/abs/2504.14572