EUVIMEDEUVIMED entdecken
Crossref

Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation

Pengcheng Xu, Boyu Wang, Charles Ling

Proceedings of the AAAI Conference on Artificial Intelligence · 26.06.2023 · 37 (3) · 3036-3044

DOI: 10.1609/aaai.v37i3.25407 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 26.06.2023. Verfasst von Pengcheng Xu und 2 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Current methods of blended targets domain adaptation (BTDA) usually infer or consider domain label information but underemphasize hybrid categorical feature structures of targets, which yields limited performance, especially under the label distribution shift. We demonstrate that domain labels are not directly necessary for BTDA if categorical distributions of various domains are sufficiently aligned even facing the imbalance of domains and the label distribution shift of classes. However, we observe that the cluster assumption in BTDA does not comprehensively hold. The hybrid categorical feature space hinders the modeling of categorical distributions and the generation of reliable pseudo labels for categorical alignment. To address these, we propose a categorical domain discriminator guided by uncertainty to explicitly model and directly align categorical distributions P(Z|Y). Simultaneously, we utilize the low-level features to augment the single source features with diverse target styles to rectify the biased classifier P(Y|Z) among diverse targets. Such a mutual conditional alignment of P(Z|Y) and P(Y|Z) forms a mutual reinforced mechanism. Our approach outperforms the state-of-the-art in BTDA even compared with methods utilizing domain labels, especially under the label distribution shift, and in single target DA on DomainNet.

Zitieren

APA

Xu, P., Wang, B., & Ling, C. (2023). Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation. Proceedings of the AAAI Conference on Artificial Intelligence, 37(3), 3036-3044. https://doi.org/10.1609/aaai.v37i3.25407

Vancouver

Xu P, Wang B, Ling C. Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation. Proceedings of the AAAI Conference on Artificial Intelligence. 2023;37(3):3036-3044. doi:10.1609/aaai.v37i3.25407

BibTeX
@article{Xu2023,
  author = {Xu, Pengcheng and Wang, Boyu and Ling, Charles},
  title = {Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2023},
  volume = {37},
  number = {3},
  pages = {3036--3044},
  doi = {10.1609/aaai.v37i3.25407}
}
Kostenlos starten →