EUVIMEDEUVIMED entdecken
Crossref

Towards Real-World Test-Time Adaptation: Tri-net Self-Training with Balanced Normalization

Yongyi Su, Xun Xu, Kui Jia

Proceedings of the AAAI Conference on Artificial Intelligence · 24.03.2024 · 38 (13) · 15126-15135

DOI: 10.1609/aaai.v38i13.29435 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 24.03.2024. Verfasst von Yongyi Su und 2 weiteren Autor:innen. Die Arbeit wurde bisher 26-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Test-Time Adaptation aims to adapt source domain model to testing data at inference stage with success demonstrated in adapting to unseen corruptions. However, these attempts may fail under more challenging real-world scenarios. Existing works mainly consider real-world test-time adaptation under non-i.i.d. data stream and continual domain shift. In this work, we first complement the existing real-world TTA protocol with a globally class imbalanced testing set. We demonstrate that combining all settings together poses new challenges to existing methods. We argue the failure of state-of-the-art methods is first caused by indiscriminately adapting normalization layers to imbalanced testing data. To remedy this shortcoming, we propose a balanced batchnorm layer to swap out the regular batchnorm at inference stage. The new batchnorm layer is capable of adapting without biasing towards majority classes. We are further inspired by the success of self-training (ST) in learning from unlabeled data and adapt ST for test-time adaptation. However, ST alone is prone to over adaption which is responsible for the poor performance under continual domain shift. Hence, we propose to improve self-training under continual domain shift by regularizing model updates with an anchored loss. The final TTA model, termed as TRIBE, is built upon a tri-net architecture with balanced batchnorm layers. We evaluate TRIBE on four datasets representing real-world TTA settings. TRIBE consistently achieves the state-of-the-art performance across multiple evaluation protocols. The code is available at https://github.com/Gorilla-Lab-SCUT/TRIBE.

Zitieren

APA

Su, Y., Xu, X., & Jia, K. (2024). Towards Real-World Test-Time Adaptation: Tri-net Self-Training with Balanced Normalization. Proceedings of the AAAI Conference on Artificial Intelligence, 38(13), 15126-15135. https://doi.org/10.1609/aaai.v38i13.29435

Vancouver

Su Y, Xu X, Jia K. Towards Real-World Test-Time Adaptation: Tri-net Self-Training with Balanced Normalization. Proceedings of the AAAI Conference on Artificial Intelligence. 2024;38(13):15126-15135. doi:10.1609/aaai.v38i13.29435

BibTeX
@article{Su2024,
  author = {Su, Yongyi and Xu, Xun and Jia, Kui},
  title = {Towards Real-World Test-Time Adaptation: Tri-net Self-Training with Balanced Normalization},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2024},
  volume = {38},
  number = {13},
  pages = {15126--15135},
  doi = {10.1609/aaai.v38i13.29435}
}
Kostenlos starten →