EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Energy-based analog neural network framework

Mohamed Watfa, Alberto Garcia-Ortiz, Gilles Sassatelli

Frontiers in Computational Neuroscience · 03.03.2023 · 17

DOI: 10.3389/fncom.2023.1114651 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Computational Neuroscience, online erschienen am 03.03.2023. Verfasst von Mohamed Watfa und 2 weiteren Autor:innen. Beteiligte Einrichtung: LIRMM, University of Montpellier, CNRS, Montpellier, France. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Over the past decade a body of work has emerged and shown the disruptive potential of neuromorphic systems across a broad range of studies, often combining novel machine learning models and nanotechnologies. Still, the scope of investigations often remains limited to simple problems since the process of building, training, and evaluating mixed-signal neural models is slow and laborious. In this paper, we introduce an open-source framework, called EBANA, that provides a unified, modularized, and extensible infrastructure, similar to conventional machine learning pipelines, for building and validating analog neural networks (ANNs). It uses Python as interface language with a syntax similar to Keras, while hiding the complexity of the underlying analog simulations. It already includes the most common building blocks and maintains sufficient modularity and extensibility to easily incorporate new concepts, electrical, and technological models. These features make EBANA suitable for researchers and practitioners to experiment with different design topologies and explore the various tradeoffs that exist in the design space. We illustrate the framework capabilities by elaborating on the increasingly popular Energy-Based Models (EBMs), used in conjunction with the local Equilibrium Propagation (EP) training algorithm. Our experiments cover 3 datasets having up to 60,000 entries and explore network topologies generating circuits in excess of 1,000 electrical nodes that can be extensively benchmarked with ease and in reasonable time thanks to the native EBANA parallelization capability.

Schlagwörter

Stichwörter der Autor:innen

SPICEanalogenergy-based modelsequilibrium propagationframeworkmixed-signalneural networks

Autor:innen und Institutionen

  • LIRMM, University of Montpellier, CNRS, Montpellier, France.
  • ITEM, University of Bremen, Bremen, Germany.

Zitieren

APA

Watfa, M., Garcia-Ortiz, A., & Sassatelli, G. (2023). Energy-based analog neural network framework. Frontiers in Computational Neuroscience, 17. https://doi.org/10.3389/fncom.2023.1114651

Vancouver

Watfa M, Garcia-Ortiz A, Sassatelli G. Energy-based analog neural network framework. Frontiers in Computational Neuroscience. 2023;17. doi:10.3389/fncom.2023.1114651

BibTeX
@article{Watfa2023,
  author = {Watfa, Mohamed and Garcia-Ortiz, Alberto and Sassatelli, Gilles},
  title = {Energy-based analog neural network framework},
  journal = {Frontiers in Computational Neuroscience},
  year = {2023},
  volume = {17},
  doi = {10.3389/fncom.2023.1114651}
}
Kostenlos starten →