Statistical Analysis of Conditional Group Distributionally Robust Optimization with Cross-Entropy Loss

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Guo, Zijian, Wang, Zhenyu, Hu, Yifan, Bach, Francis
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908783587885056
author Guo, Zijian
Wang, Zhenyu
Hu, Yifan
Bach, Francis
author_facet Guo, Zijian
Wang, Zhenyu
Hu, Yifan
Bach, Francis
contents In multi-source learning with discrete labels, distributional heterogeneity across domains poses a central challenge to developing predictive models that transfer reliably to unseen domains. We study multi-source unsupervised domain adaptation, where labeled data are available from multiple source domains and only unlabeled data are observed from the target domain. To address potential distribution shifts, we propose a novel Conditional Group Distributionally Robust Optimization (CG-DRO) framework that learns a classifier by minimizing the worst-case cross-entropy loss over the convex combinations of the conditional outcome distributions from sources domains. We develop an efficient Mirror Prox algorithm for solving the minimax problem and employ a double machine learning procedure to estimate the risk function, ensuring that errors in nuisance estimation contribute only at higher-order rates. We establish fast statistical convergence rates for the empirical CG-DRO estimator by constructing two surrogate minimax optimization problems that serve as theoretical bridges. A distinguishing challenge for CG-DRO is the emergence of nonstandard asymptotics: the empirical CG-DRO estimator may fail to converge to a standard limiting distribution due to boundary effects and system instability. To address this, we introduce a perturbation-based inference procedure that enables uniformly valid inference, including confidence interval construction and hypothesis testing.
format Preprint
id arxiv_https___arxiv_org_abs_2507_09905
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Statistical Analysis of Conditional Group Distributionally Robust Optimization with Cross-Entropy Loss
Guo, Zijian
Wang, Zhenyu
Hu, Yifan
Bach, Francis
Methodology
Optimization and Control
Statistics Theory
Machine Learning
In multi-source learning with discrete labels, distributional heterogeneity across domains poses a central challenge to developing predictive models that transfer reliably to unseen domains. We study multi-source unsupervised domain adaptation, where labeled data are available from multiple source domains and only unlabeled data are observed from the target domain. To address potential distribution shifts, we propose a novel Conditional Group Distributionally Robust Optimization (CG-DRO) framework that learns a classifier by minimizing the worst-case cross-entropy loss over the convex combinations of the conditional outcome distributions from sources domains. We develop an efficient Mirror Prox algorithm for solving the minimax problem and employ a double machine learning procedure to estimate the risk function, ensuring that errors in nuisance estimation contribute only at higher-order rates. We establish fast statistical convergence rates for the empirical CG-DRO estimator by constructing two surrogate minimax optimization problems that serve as theoretical bridges. A distinguishing challenge for CG-DRO is the emergence of nonstandard asymptotics: the empirical CG-DRO estimator may fail to converge to a standard limiting distribution due to boundary effects and system instability. To address this, we introduce a perturbation-based inference procedure that enables uniformly valid inference, including confidence interval construction and hypothesis testing.
title Statistical Analysis of Conditional Group Distributionally Robust Optimization with Cross-Entropy Loss
topic Methodology
Optimization and Control
Statistics Theory
Machine Learning
url https://arxiv.org/abs/2507.09905