Domain-Free Adversarial Splitting for Domain Generalization

Domain generalization is an approach that utilizes several source domains to train the learner to be generalizable to unseen target domain to tackle domain shift issue. It has drawn much attention in machine learning community. This paper aims to learn to generalize well to unseen target domain without relying on the knowledge of the number of source domains and domain labels. To achieve that goal, we unify adversarial training and meta-learning in a novel proposed Domain-Free Adversarial Splitting (DFAS) framework. In this framework, we model the domain generalization as a learning problem that enforces the learner to be able to generalize well for any train/val subsets splitting of the training dataset. This model can be further transformed to be a min-max optimization problem which can be solved by an iterative adversarial training process. In each iteration, it adversarially splits the training dataset into train/val subsets to maximize domain shift between them using current learner, and then updates the learner on this splitting to be able to generalize well from train-subset to val-subset using meta-learning approach. Extensive experiments on three benchmark datasets under three different settings on the source and target domains show that our method achieves state-of-the-art results and confirm the effectiveness of our method by ablation study. We also derive a generalization error bound for theoretical understanding of our method.