Domain Invariant Representation Learning with Domain Density Transformations

Abstract

Domain generalization is the problem where we aim to train a model with data from a set of source domains so that the model can generalize to unseen target domains. Naively training a model on the aggregate set of data (pooled from all source domains) has been shown to perform poorly, since the information learned by that model might be domain-specific and cannot generalize well to target domains. To tackle this problem, a predominant approach is to find and learn some domain-invariant information and use that for the prediction problem. In this paper, we propose a theoretically grounded method to learn a domain-invariant representation by enforcing the representation network to be invariant under all transformation functions among domains. We also show how to use Generative Adversarial Networks to learn such domain transformations to implement our method in practice. We illustrate the effectiveness of our method on several widely used datasets for domain generalization problem, on all of which we achieve competitive results with state-of-the-art models.