arXiv Analytics

Sign in

arXiv:2010.07922 [cs.LG]AbstractReferencesReviewsResources

Representation Learning via Invariant Causal Mechanisms

Jovana Mitrovic, Brian McWilliams, Jacob Walker, Lars Buesing, Charles Blundell

Published 2020-10-15Version 1

Self-supervised learning has emerged as a strategy to reduce the reliance on costly supervised signal by pretraining representations only using unlabeled data. These methods combine heuristic proxy classification tasks with data augmentations and have achieved significant success, but our theoretical understanding of this success remains limited. In this paper we analyze self-supervised representation learning using a causal framework. We show how data augmentations can be more effectively utilized through explicit invariance constraints on the proxy classifiers employed during pretraining. Based on this, we propose a novel self-supervised objective, Representation Learning via Invariant Causal Mechanisms (ReLIC), that enforces invariant prediction of proxy targets across augmentations through an invariance regularizer which yields improved generalization guarantees. Further, using causality we generalize contrastive learning, a particular kind of self-supervised method, and provide an alternative theoretical explanation for the success of these methods. Empirically, ReLIC significantly outperforms competing methods in terms of robustness and out-of-distribution generalization on ImageNet, while also significantly outperforming these methods on Atari achieving above human-level performance on $51$ out of $57$ games.

Related articles: Most relevant | Search more
arXiv:1909.09252 [cs.LG] (Published 2019-09-19)
HyperLearn: A Distributed Approach for Representation Learning in Datasets With Many Modalities
arXiv:2311.08815 [cs.LG] (Published 2023-11-15)
Self-Supervised Disentanglement by Leveraging Structure in Data Augmentations
arXiv:2206.11646 [cs.LG] (Published 2022-06-23)
Invariant Causal Mechanisms through Distribution Matching