Repeated Environment Inference for Invariant Learning

TL;DR

This paper introduces an iterative environment inference method that improves invariant learning by repeatedly refining environment partitions, leading to better spurious correlation capture and overall performance.

Contribution

It proposes a novel repeated environment inference approach that enhances invariant representation learning without known environment labels.

Findings

Outperforms baseline methods on synthetic datasets.

Achieves superior results on real-world datasets.

Iterative process better captures spurious correlations.

Abstract

We study the problem of invariant learning when the environment labels are unknown. We focus on the invariant representation notion when the Bayes optimal conditional label distribution is the same across different environments. Previous work conducts Environment Inference (EI) by maximizing the penalty term from Invariant Risk Minimization (IRM) framework. The EI step uses a reference model which focuses on spurious correlations to efficiently reach a good environment partition. However, it is not clear how to find such a reference model. In this work, we propose to repeat the EI process and retrain an ERM model on the \textit{majority} environment inferred by the previous EI step. Under mild assumptions, we find that this iterative process helps learn a representation capturing the spurious correlation better than the single step. This results in better Environment Inference and better Invariant Learning. We show that this method outperforms baselines on both synthetic and real-world datasets.