EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Bringing Equity to Classification: Domain Generalization for Domain-Linked Classes

Kimathi Kaai, Saad Hossain, Sirisha Rambhatla

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2025-11 · 47 (11) · 9894-9905

DOI: 10.1109/tpami.2025.3593407 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025-11. Verfasst von Kimathi Kaai und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 87 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Sciences and Engineering Research Council of Canada.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Domain generalization (DG) focuses on transferring domain-invariant knowledge from multiple source (training) domains to an a priori unseen target domain(s). This task implicitly requires that classes of interest are expressed in multiple sources (domain-shared) to break spurious domain-class correlations. However, real-world data scarcity challenges may often result in classes present in only a specific domain (domain-linked), which we show leads to extremely poor generalization. In this work, we introduce the domain-linked DG task to the community and develop a methodology to learn useful domain-invariant representations from domain-shared classes for domain-linked ones. Specifically, we propose FOND, a Fairness-inspired and cONtrastive learning objective for Domain-linked DG. Rigorous and reproducible experimental results communicate that FOND accomplishes state-of-the-art improvements for domain-linked classes, given a sufficient number of domain-shared classes and with minimal performance trade-offs. Complementary to these contributions, we theoretically analyze this task and provide practical insights for domain-linked class generalizability.

Förderung

  • Natural Sciences and Engineering Research Council of Canada

Zitieren

APA

Kaai, K., Hossain, S., & Rambhatla, S. (2025). Bringing Equity to Classification: Domain Generalization for Domain-Linked Classes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 47(11), 9894-9905. https://doi.org/10.1109/tpami.2025.3593407

Vancouver

Kaai K, Hossain S, Rambhatla S. Bringing Equity to Classification: Domain Generalization for Domain-Linked Classes. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2025;47(11):9894-9905. doi:10.1109/tpami.2025.3593407

BibTeX
@article{Kaai2025,
  author = {Kaai, Kimathi and Hossain, Saad and Rambhatla, Sirisha},
  title = {Bringing Equity to Classification: Domain Generalization for Domain-Linked Classes},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2025},
  volume = {47},
  number = {11},
  pages = {9894--9905},
  doi = {10.1109/tpami.2025.3593407}
}
Kostenlos starten →