EUVIMEDEUVIMED entdecken
Crossref

Backpropagation-free training of deep physical neural networks

Ali Momeni iD, Babak Rahmani iD, Matthieu Malléjac iD, Philipp Del Hougne iD, Romain Fleury iD

Science · 15.12.2023 · 382 (6676) · 1297-1303

DOI: 10.1126/science.adi8474 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science, erschienen am 15.12.2023. Verfasst von Ali Momeni und 4 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Wave Engineering, Department of Electrical Engineering, EPFL, Lausanne CH-1015, Switzerland. Die Arbeit wurde bisher 140-mal zitiert (Crossref) und nennt 69 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent successes in deep learning for vision and natural language processing are attributed to larger models but come with energy consumption and scalability issues. Current training of digital deep-learning models primarily relies on backpropagation that is unsuitable for physical implementation. In this work, we propose a simple deep neural network architecture augmented by a physical local learning (PhyLL) algorithm, which enables supervised and unsupervised training of deep physical neural networks without detailed knowledge of the nonlinear physical layer’s properties. We trained diverse wave-based physical neural networks in vowel and image classification experiments, showcasing the universality of our approach. Our method shows advantages over other hardware-aware training schemes by improving training speed, enhancing robustness, and reducing power consumption by eliminating the need for system modeling and thus decreasing digital computation.

Autor:innen und Institutionen

  • Laboratory of Wave Engineering, Department of Electrical Engineering, EPFL, Lausanne CH-1015, Switzerland.
  • Microsoft Research, Cambridge CB4 0AB, UK.
  • University of Rennes, CNRS, IETR - UMR 6164, F-35000 Rennes, France.

ORCID-Profile: Ali Momeni, Babak Rahmani, Matthieu Malléjac, Philipp Del Hougne, Romain Fleury

Zitieren

APA

Momeni, A., Rahmani, B., Malléjac, M., Del Hougne, P., & Fleury, R. (2023). Backpropagation-free training of deep physical neural networks. Science, 382(6676), 1297-1303. https://doi.org/10.1126/science.adi8474

Vancouver

Momeni A, Rahmani B, Malléjac M, Del Hougne P, Fleury R. Backpropagation-free training of deep physical neural networks. Science. 2023;382(6676):1297-1303. doi:10.1126/science.adi8474

BibTeX
@article{Momeni2023,
  author = {Momeni, Ali and Rahmani, Babak and Malléjac, Matthieu and Del Hougne, Philipp and Fleury, Romain},
  title = {Backpropagation-free training of deep physical neural networks},
  journal = {Science},
  year = {2023},
  volume = {382},
  number = {6676},
  pages = {1297--1303},
  doi = {10.1126/science.adi8474}
}
Kostenlos starten →