Learning Under Adversarial and Interventional Shifts

TL;DR

This paper introduces RISe, a new approach that combines adversarial and interventional shift robustness to create models resilient to a broader range of distribution changes, validated through experiments in healthcare.

Contribution

The paper proposes RISe, a novel formulation that integrates adversarial and interventional shifts for robust model training, using distributionally robust optimization in supervised and reinforcement learning.

Findings

Effective robustness against combined distribution shifts demonstrated

Improved performance on healthcare datasets with synthetic and real data

Outperforms existing methods in robustness benchmarks

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.