Contrastive ACE: Domain Generalization Through Alignment of Causal Mechanisms

TL;DR

This paper introduces a novel domain generalization method called Contrastive ACE that enforces invariance of causal mechanisms across domains by aligning the average causal effect of features, improving generalization to unseen domains.

Contribution

It proposes a new causal invariance regularization based on the average causal effect, enhancing domain generalization by aligning causal mechanisms across domains.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Demonstrates the effectiveness of causal mechanism alignment.

Shows improved stability of causal predictions across domains.

Abstract

Domain generalization aims to learn knowledge invariant across different distributions while semantically meaningful for downstream tasks from multiple source domains, to improve the model's generalization ability on unseen target domains. The fundamental objective is to understand the underlying "invariance" behind these observational distributions and such invariance has been shown to have a close connection to causality. While many existing approaches make use of the property that causal features are invariant across domains, we consider the causal invariance of the average causal effect of the features to the labels. This invariance regularizes our training approach in which interventions are performed on features to enforce stability of the causal prediction by the classifier across domains. Our work thus sheds some light on the domain generalization problem by introducing invariance of the mechanisms into the learning process. Experiments on several benchmark datasets demonstrate the performance of the proposed method against SOTAs.