How Does Distribution Matching Help Domain Generalization: An Information-Theoretic Analysis
IEEE Transactions on Information Theory · 2025-03 · 71 (3) · 2028-2053
DOI: 10.1109/tit.2025.3531136 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Information Theory, 2025-03. Verfasst von Yuxin Dong und 5 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 83 Literaturangaben. Gefördert von: Key Research and Development Project in Shaanxi Province, National Natural Science Foundation of China.
Abstract (Originalsprache)
Domain generalization aims to learn invariance across multiple source domains, thereby enhancing generalization against out-of-distribution data. While gradient or representation matching algorithms have achieved remarkable success in domain generalization, these methods generally lack generalization guarantees or depend on strong assumptions, leaving a gap in understanding the underlying mechanism of distribution matching. In this work, we formulate domain generalization from a novel probabilistic perspective, ensuring robustness while avoiding overly conservative solutions. Through comprehensive information-theoretic analysis, we provide key insights into the roles of gradient and representation matching in promoting generalization. Our results reveal the complementary relationship between these two components, indicating that existing works focusing solely on either gradient or representation alignment are insufficient to solve the domain generalization problem. In light of these theoretical findings, we introduce IDM to simultaneously align the inter-domain gradients and representations. Integrated with the proposed PDM method for complex distribution matching, IDM achieves superior performance over various baseline methods.
Förderung
- Key Research and Development Project in Shaanxi Province
- National Natural Science Foundation of China
Zitieren
APA
Dong, Y., Gong, T., Chen, H., Song, S., Zhang, W., & Li, C. (2025). How Does Distribution Matching Help Domain Generalization: An Information-Theoretic Analysis. IEEE Transactions on Information Theory, 71(3), 2028-2053. https://doi.org/10.1109/tit.2025.3531136
Vancouver
Dong Y, Gong T, Chen H, Song S, Zhang W, Li C. How Does Distribution Matching Help Domain Generalization: An Information-Theoretic Analysis. IEEE Transactions on Information Theory. 2025;71(3):2028-2053. doi:10.1109/tit.2025.3531136
BibTeX
@article{Dong2025,
author = {Dong, Yuxin and Gong, Tieliang and Chen, Hong and Song, Shuangyong and Zhang, Weizhan and Li, Chen},
title = {How Does Distribution Matching Help Domain Generalization: An Information-Theoretic Analysis},
journal = {IEEE Transactions on Information Theory},
year = {2025},
volume = {71},
number = {3},
pages = {2028--2053},
doi = {10.1109/tit.2025.3531136}
}
EUVIMED