Domain Generalization Using a Mixture of Multiple Latent Domains

TL;DR

This paper proposes a novel domain generalization approach that automatically discovers latent domains through style-based clustering and adversarial training, enabling improved performance on unseen domains without requiring domain labels.

Contribution

Introduces a method for domain generalization that does not rely on domain labels by clustering based on style features and adversarial learning to identify latent domains.

Findings

Successfully discovers latent domains without labels

Outperforms conventional domain generalization methods

Enhances model performance on unseen target domains

Abstract

When domains, which represent underlying data distributions, vary during training and testing processes, deep neural networks suffer a drop in their performance. Domain generalization allows improvements in the generalization performance for unseen target domains by using multiple source domains. Conventional methods assume that the domain to which each sample belongs is known in training. However, many datasets, such as those collected via web crawling, contain a mixture of multiple latent domains, in which the domain of each sample is unknown. This paper introduces domain generalization using a mixture of multiple latent domains as a novel and more realistic scenario, where we try to train a domain-generalized model without using domain labels. To address this scenario, we propose a method that iteratively divides samples into latent domains via clustering, and which trains the domain-invariant feature extractor shared among the divided latent domains via adversarial learning. We assume that the latent domain of images is reflected in their style, and thus, utilize style features for clustering. By using these features, our proposed method successfully discovers latent domains and achieves domain generalization even if the domain labels are not given. Experiments show that our proposed method can train a domain-generalized model without using domain labels. Moreover, it outperforms conventional domain generalization methods, including those that utilize domain labels.