On the Connection between Invariant Learning and Adversarial Training for Out-of-Distribution Generalization

Open in new window