English

Learning Under Adversarial and Interventional Shifts

Machine Learning 2021-03-31 v1 Machine Learning

Abstract

Machine learning models are often trained on data from one distribution and deployed on others. So it becomes important to design models that are robust to distribution shifts. Most of the existing work focuses on optimizing for either adversarial shifts or interventional shifts. Adversarial methods lack expressivity in representing plausible shifts as they consider shifts to joint distributions in the data. Interventional methods allow more expressivity but provide robustness to unbounded shifts, resulting in overly conservative models. In this work, we combine the complementary strengths of the two approaches and propose a new formulation, RISe, for designing robust models against a set of distribution shifts that are at the intersection of adversarial and interventional shifts. We employ the distributionally robust optimization framework to optimize the resulting objective in both supervised and reinforcement learning settings. Extensive experimentation with synthetic and real world datasets from healthcare demonstrate the efficacy of the proposed approach.

Keywords

Cite

@article{arxiv.2103.15933,
  title  = {Learning Under Adversarial and Interventional Shifts},
  author = {Harvineet Singh and Shalmali Joshi and Finale Doshi-Velez and Himabindu Lakkaraju},
  journal= {arXiv preprint arXiv:2103.15933},
  year   = {2021}
}

Comments

19 pages including 5 pages appendix, 6 figures, 2 tables. Preliminary version presented at Causal Discovery & Causality-Inspired Machine Learning Workshop 2020

R2 v1 2026-06-24T00:40:05.492Z