Invariant Causal Mechanisms through Distribution Matching

Mathieu Chevalley, Charlotte Bunne, Andreas Krause, Stefan Bauer

Keywords: representation learning, causality, invariance, distribution matching

Abstract: Learning representations that capture the underlying data generating process is akey problem for data efficient and robust use of neural networks. One key property for robustness which the learned representation should capture and which recently received a lot of attention is described by the notion of invariance. In this work we provide a causal perspective and new algorithm for learning invariant representations. Empirically we show that this algorithm works well on a diverse set of tasks and in particular we observe state-of-the-art performance on domain generalization, where we are able to significantly boost the score of existing models.

One-sentence Summary: This work provides a causal perspective on new algorithm for invariant representation learning.