Invariant Risk Minimization
This addresses the challenge of building robust machine learning models that perform reliably across diverse environments, representing a foundational advance in causal inference and generalization.
The authors tackled the problem of out-of-distribution generalization by introducing Invariant Risk Minimization (IRM), a learning paradigm that estimates invariant correlations across multiple training distributions, enabling predictions that generalize beyond the training data.
We introduce Invariant Risk Minimization (IRM), a learning paradigm to estimate invariant correlations across multiple training distributions. To achieve this goal, IRM learns a data representation such that the optimal classifier, on top of that data representation, matches for all training distributions. Through theory and experiments, we show how the invariances learned by IRM relate to the causal structures governing the data and enable out-of-distribution generalization.