EUVIMEDEUVIMED entdecken
Crossref

Client-level Active Error Correction in Distributed Learning

Kwang In Kim

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (27) · 22591-22599

DOI: 10.1609/aaai.v40i27.39419 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Kwang In Kim.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Label errors can significantly degrade model performance, making effective mechanisms crucial. Active error correction (AEC) addresses this by prioritizing data points for human re-labeling where corrections are expected to have significant impact. We extend AEC to distributed collaborative learning, where clients hold local data and a central server allocates labeling resources. Existing AEC methods assume centralized access and do not generalize to distributed settings. To overcome this, we use neural network weight gradients from client updates as proxies for local data and apply a Gaussian process in gradient space to strategically select clients for correction. Our method identifies gradient inconsistencies and encourages diversity through a computationally efficient rank-one Cholesky update. Experiments on eight benchmark datasets demonstrate the effectiveness of our approach.

Zitieren

APA

Kim, K. I. (2026). Client-level Active Error Correction in Distributed Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(27), 22591-22599. https://doi.org/10.1609/aaai.v40i27.39419

Vancouver

Kim KI. Client-level Active Error Correction in Distributed Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(27):22591-22599. doi:10.1609/aaai.v40i27.39419

BibTeX
@article{Kim2026,
  author = {Kim, Kwang In},
  title = {Client-level Active Error Correction in Distributed Learning},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {27},
  pages = {22591--22599},
  doi = {10.1609/aaai.v40i27.39419}
}
Kostenlos starten →