Invariant Learning with Annotation-free Environments

Phuong Quynh Le, Christin Seifert, Jörg Schlötterer

Track: Extended Abstract Track

Keywords: invariant learning, spurious correlations

TL;DR: This paper proposes an efficient approach to obtaining environments beneficial for invariant learning without additional annotations.

Abstract: Invariant learning across environments is a promising approach to improve domain generalization compared to Empirical Risk Minimization (ERM). However, most invariant learning methods rely on the assumption that training examples are pre-partitioned into different known environments. We instead infer environments without the need for additional annotations, motivated by observations of the properties within the representation space of a trained ERM model. We show the preliminary effectiveness of our approach on the ColoredMNIST benchmark, achieving performance comparable to methods requiring explicit environment labels and on par with an annotation-free method that poses strong restrictions on the ERM reference model.

Submission Number: 48