Linear unit-tests for invariance discovery

Abstract

There is an increasing interest in algorithms to learn invariant correlations across training environments. A big share of the current proposals find theoretical support in the causality literature but, how useful are they in practice? The purpose of this note is to propose six linear low-dimensional problems -- unit tests -- to evaluate different types of out-of-distribution generalization in a precise manner. Following initial experiments, none of the three recently proposed alternatives passes all tests. By providing the code to automatically replicate all the results in this manuscript (https://www.github.com/facebookresearch/InvarianceUnitTests), we hope that our unit tests become a standard steppingstone for researchers in out-of-distribution generalization.

5 pages, Causal Discovery & Causality-Inspired Machine Learning Workshop at Neural Information Processing Systems

References in corpus (5)

• Scikit-learn: Machine Learning in Python
• Shortcut Learning in Deep Neural Networks
• Invariant Risk Minimization
• Learning explanations that are hard to vary
• When is invariance useful in an Out-of-Distribution Generalization problem ?

Cited by in corpus (7)

• Towards Out-Of-Distribution Generalization: A Survey
• Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization
• When is invariance useful in an Out-of-Distribution Generalization problem ?
• Iterative Feature Matching: Toward Provable Domain Generalization with Logarithmic Environments
• On Invariance Penalties for Risk Minimization
• Invariant Risk Minimisation for Cross-Organism Inference: Substituting Mouse Data for Human Data in Human Risk Factor Discovery
• Beyond Discriminant Patterns: On the Robustness of Decision Rule Ensembles