EUVIMEDEUVIMED entdecken
Crossref

Adaptive adapter routing for long-tailed class-incremental learning

Zhi-Hong Qi, Da-Wei Zhou, Yiran Yao, Han-Jia Ye, De-Chuan Zhan

Machine Learning · 2025-03 · 114 (3)

DOI: 10.1007/s10994-024-06662-4 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Machine Learning, online erschienen am 06.02.2025. Verfasst von Zhi-Hong Qi und 4 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 44 Literaturangaben. Gefördert von: National Science and Technology Major Project, NSFC, Collaborative Innovation Center of Novel Software Technology and Industrialization.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In our ever-evolving world, new data exhibits a long-tailed distribution, such as emerging images in varying amounts. This necessitates continuous model learning imbalanced data without forgetting, addressing the challenge of long-tailed class-incremental learning (LTCIL). Existing methods often rely on retraining linear classifiers with former data, which is impractical in real-world settings. In this paper, we harness the potent representation capabilities of pre-trained models and introduce AdaPtive Adapter RouTing (Apart) as an exemplar-free solution for LTCIL. To counteract forgetting, we train inserted adapters with frozen pre-trained weights for deeper adaptation and maintain a pool of adapters for selection during sequential model updates. Additionally, we present an auxiliary adapter pool designed for effective generalization, especially on minority classes. Adaptive instance routing across these pools captures crucial correlations, facilitating a comprehensive representation of all classes. Consequently, Apart tackles the imbalance problem as well as catastrophic forgetting in a unified framework. Extensive benchmark experiments validate the effectiveness of Apart.

Förderung

  • National Science and Technology Major Project
  • NSFC
  • Collaborative Innovation Center of Novel Software Technology and Industrialization

Zitieren

APA

Qi, Z., Zhou, D., Yao, Y., Ye, H., & Zhan, D. (2025). Adaptive adapter routing for long-tailed class-incremental learning. Machine Learning, 114(3). https://doi.org/10.1007/s10994-024-06662-4

Vancouver

Qi Z, Zhou D, Yao Y, Ye H, Zhan D. Adaptive adapter routing for long-tailed class-incremental learning. Machine Learning. 2025;114(3). doi:10.1007/s10994-024-06662-4

BibTeX
@article{Qi2025,
  author = {Qi, Zhi-Hong and Zhou, Da-Wei and Yao, Yiran and Ye, Han-Jia and Zhan, De-Chuan},
  title = {Adaptive adapter routing for long-tailed class-incremental learning},
  journal = {Machine Learning},
  year = {2025},
  volume = {114},
  number = {3},
  doi = {10.1007/s10994-024-06662-4}
}
Kostenlos starten →