Domain Generalization using Causal Matching

Abstract

In the domain generalization literature, a common objective is to learn representations independent of the domain after conditioning on the class label. We show that this objective is not sufficient: there exist counter-examples where a model fails to generalize to unseen domains even after satisfying class-conditional domain invariance. We formalize this observation through a structural causal model and show the importance of modeling within-class variations for generalization. Specifically, classes contain objects that characterize specific causal features, and domains can be interpreted as interventions on these objects that change non-causal features. We highlight an alternative condition: inputs across domains should have the same representation if they are derived from the same object. Based on this objective, we propose matching-based algorithms when base objects are observed (e.g., through data augmentation) and approximate the objective when objects are not observed (MatchDG). Our simple matching-based algorithms are competitive to prior work on out-of-domain accuracy for rotated MNIST, Fashion-MNIST, PACS, and Chest-Xray datasets. Our method MatchDG also recovers ground-truth object matches: on MNIST and Fashion-MNIST, top-10 matches from MatchDG have over 50% overlap with ground-truth matches.

arXiv commentProceedings of the 38th International Conference on Machine Learning (ICML), PMLR 139, 2021. (Long Talk)

References in corpus (26)

• A Simple Framework for Contrastive Learning of Visual Representations
• Momentum Contrast for Unsupervised Visual Representation Learning
• Domain Generalization via Invariant Feature Representation
• Contrastive Multiview Coding
• Domain Generalization via Model-Agnostic Learning of Semantic Features
• Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization
• Generalizing Across Domains via Cross-Gradient Training
• Invariant Risk Minimization
• Out-of-Distribution Generalization via Risk Extrapolation (REx)
• On Learning Invariant Representation for Domain Adaptation
• Invariant Models for Causal Transfer Learning
• Improve Unsupervised Domain Adaptation with Mixup Training
• Heterogeneous Domain Generalization via Domain Mixup
• Learning Robust Representations by Projecting Superficial Statistics Out
• Generalizing to unseen domains via distribution matching
• The Pitfalls of Simplicity Bias in Neural Networks
• In Search of Lost Domain Generalization
• Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributions
• On the limits of cross-domain generalization in automated X-ray prediction
• Efficient Domain Generalization via Common-Specific Low-Rank Decomposition
• Domain Generalization via Multidomain Discriminant Analysis
• Invariant Risk Minimization Games
• Domain Generalization via Conditional Invariant Representation
• Improving out-of-distribution generalization via multi-task self-supervised pretraining
• Towards Shape Biased Unsupervised Representation Learning for Domain Generalization
• Sequential Learning for Domain Generalization

Cited by in corpus (17)

• Domain Generalization: A Survey
• The Risks of Invariant Risk Minimization
• Generalizing to unseen domains via distribution matching
• Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization
• Hierarchical Disentanglement-Alignment Network for Robust SAR Vehicle Recognition
• Towards out of distribution generalization for problems in mechanics
• Iterative Feature Matching: Toward Provable Domain Generalization with Logarithmic Environments
• Towards Causal Federated Learning For Enhanced Robustness and Privacy
• An Information-theoretic Approach to Distribution Shifts
• Alleviating Privacy Attacks via Causal Learning
• Hierarchical Variational Auto-Encoding for Unsupervised Domain Generalization
• Generalizing Nucleus Recognition Model in Multi-source Images via Pruning
• Explainability-aided Domain Generalization for Image Classification
• Improved Detection of Face Presentation Attacks Using Image Decomposition
• Towards Data-Free Domain Generalization
• Matched sample selection with GANs for mitigating attribute confounding
• Deep Domain Generalization with Feature-norm Network