Domain Generalization through the Lens of Angular Invariance

TL;DR

This paper introduces a novel domain generalization method called AIDGN that leverages angular invariance and norm shift assumptions, relaxing traditional distribution alignment requirements, and demonstrates its effectiveness on multiple benchmarks.

Contribution

The paper proposes a new angular invariance principle and a deep DG method, AIDGN, using a von-Mises Fisher mixture model to improve domain generalization.

Findings

AIDGN outperforms existing methods on multiple DG benchmarks.

Angular invariance provides a robust alternative to distribution alignment.

The method effectively handles domain shifts in real-world scenarios.

Abstract

Domain generalization (DG) aims at generalizing a classifier trained on multiple source domains to an unseen target domain with domain shift. A common pervasive theme in existing DG literature is domain-invariant representation learning with various invariance assumptions. However, prior works restrict themselves to a radical assumption for realworld challenges: If a mapping induced by a deep neural network (DNN) could align the source domains well, then such a mapping aligns a target domain as well. In this paper, we simply take DNNs as feature extractors to relax the requirement of distribution alignment. Specifically, we put forward a novel angular invariance and the accompanied norm shift assumption. Based on the proposed term of invariance, we propose a novel deep DG method called Angular Invariance Domain Generalization Network (AIDGN). The optimization objective of AIDGN is developed with a von-Mises Fisher (vMF) mixture model. Extensive experiments on multiple DG benchmark datasets validate the effectiveness of the proposed AIDGN method.