Compound Domain Generalization via Meta-Knowledge Encoding

TL;DR

This paper introduces COMEN, a novel method for compound domain generalization that automatically discovers latent domains and models holistic semantic structures to improve generalization to unseen domains.

Contribution

COMEN proposes a two-step approach with SDNorm for latent domain discovery and prototype-based relational modeling to encode meta-knowledge without domain labels.

Findings

COMEN outperforms state-of-the-art methods on four DG benchmarks.

It effectively discovers latent domains without explicit domain supervision.

The approach enhances out-of-distribution generalization through semantic structure encoding.

Abstract

Domain generalization (DG) aims to improve the generalization performance for an unseen target domain by using the knowledge of multiple seen source domains. Mainstream DG methods typically assume that the domain label of each source sample is known a priori, which is challenged to be satisfied in many real-world applications. In this paper, we study a practical problem of compound DG, which relaxes the discrete domain assumption to the mixed source domains setting. On the other hand, current DG algorithms prioritize the focus on semantic invariance across domains (one-vs-one), while paying less attention to the holistic semantic structure (many-vs-many). Such holistic semantic structure, referred to as meta-knowledge here, is crucial for learning generalizable representations. To this end, we present Compound Domain Generalization via Meta-Knowledge Encoding (COMEN), a general approach to automatically discover and model latent domains in two steps. Firstly, we introduce Style-induced Domain-specific Normalization (SDNorm) to re-normalize the multi-modal underlying distributions, thereby dividing the mixture of source domains into latent clusters. Secondly, we harness the prototype representations, the centroids of classes, to perform relational modeling in the embedding space with two parallel and complementary modules, which explicitly encode the semantic structure for the out-of-distribution generalization. Experiments on four standard DG benchmarks reveal that COMEN exceeds the state-of-the-art performance without the need of domain supervision.