EUVIMEDEUVIMED entdecken
Crossref

Training Neural Networks at Any Scale: An exposition

Thomas Pethick, Kimon Antonakopoulos, Antonio Silveti-Falls, Leena Chennuru Vankadara, Volkan Cevher

IEEE Signal Processing Magazine · 2026-05 · 43 (3) · 21-36

DOI: 10.1109/msp.2026.3667310 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Signal Processing Magazine, 2026-05. Verfasst von Thomas Pethick und 4 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Gefördert von: Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, Army Research Office.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This article reviews modern optimization methods for training neural networks (NNs) with an emphasis on efficiency and scale. We present state-of-the-art optimization algorithms under a unified algorithmic template that highlights the importance of adapting to the structures in the problem. We then cover how to make these algorithms agnostic to the scale of the problem. Our exposition is intended as an introduction for both practitioners and researchers who wish to be involved in these exciting new developments.

Förderung

  • Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung
  • Army Research Office

Zitieren

APA

Pethick, T., Antonakopoulos, K., Silveti-Falls, A., Vankadara, L. C., & Cevher, V. (2026). Training Neural Networks at Any Scale: An exposition. IEEE Signal Processing Magazine, 43(3), 21-36. https://doi.org/10.1109/msp.2026.3667310

Vancouver

Pethick T, Antonakopoulos K, Silveti-Falls A, Vankadara LC, Cevher V. Training Neural Networks at Any Scale: An exposition. IEEE Signal Processing Magazine. 2026;43(3):21-36. doi:10.1109/msp.2026.3667310

BibTeX
@article{Pethick2026,
  author = {Pethick, Thomas and Antonakopoulos, Kimon and Silveti-Falls, Antonio and Vankadara, Leena Chennuru and Cevher, Volkan},
  title = {Training Neural Networks at Any Scale: An exposition},
  journal = {IEEE Signal Processing Magazine},
  year = {2026},
  volume = {43},
  number = {3},
  pages = {21--36},
  doi = {10.1109/msp.2026.3667310}
}
Kostenlos starten →