Event-driven adaptive optical neural network
Science Advances · 20.10.2023 · 9 (42)
DOI: 10.1126/sciadv.adi9127 ↗Kurz gefasst
Fachartikel in der Zeitschrift Science Advances, erschienen am 20.10.2023. Verfasst von Frank Brückerhoff-Plückelmann und 11 weiteren Autor:innen. Beteiligte Einrichtung: Physical Institute, University of Münster, Heisenbergstraße 11, 48149 Münster, Germany. Die Arbeit wurde bisher 51-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
We present an adaptive optical neural network based on a large-scale event-driven architecture. In addition to changing the synaptic weights (synaptic plasticity), the optical neural network’s structure can also be reconfigured enabling various functionalities (structural plasticity). Key building blocks are wavelength-addressable artificial neurons with embedded phase-change materials that implement nonlinear activation functions and nonvolatile memory. Using multimode focusing, the activation function features both excitatory and inhibitory responses and shows a reversible switching contrast of 3.2 decibels. We train the neural network to distinguish between English and German text samples via an evolutionary algorithm. We investigate both the synaptic and structural plasticity during the training process. On the basis of this concept, we realize a large-scale network consisting of 736 subnetworks with 16 phase-change material neurons each. Overall, 8398 neurons are functional, highlighting the scalability of the photonic architecture.
Autor:innen und Institutionen
- Physical Institute, University of Münster, Heisenbergstraße 11, 48149 Münster, Germany.
- Institute for Geoinformatics, University of Münster, Heisenbergstraße 2, 48149 Münster, Germany.
- Institute of Materials Physics, University of Münster, Wilhelm-Klemm-Straße 10, 48149 Münster, Germany.
- Department of Material, University of Oxford, Parks Road, Oxford OX1 3PH, UK.
- Department of Engineering, University of Exeter, North Park Road, Exeter EX4 4QF, UK.
- Kirchhoff-Institute for Physics, University of Heidelberg, Im Neuenheimer Feld 227, 69120 Heidelberg, Germany.
ORCID-Profile: Frank Brückerhoff-Plückelmann, Ivonne Bente, Nikolaos Farmakidis, Emma Lomonte, Francesco Lenzini, C David Wright, Harish Bhaskaran, Martin Salinga, Benjamin Risse, Wolfram H P Pernice
Zitieren
APA
Brückerhoff-Plückelmann, F., Bente, I., Becker, M., Vollmar, N., Farmakidis, N., Lomonte, E., Lenzini, F., Wright, C. D., Bhaskaran, H., Salinga, M., Risse, B., & Pernice, W. H. P. (2023). Event-driven adaptive optical neural network. Science Advances, 9(42). https://doi.org/10.1126/sciadv.adi9127
Vancouver
Brückerhoff-Plückelmann F, Bente I, Becker M, Vollmar N, Farmakidis N, Lomonte E, et al. Event-driven adaptive optical neural network. Science Advances. 2023;9(42). doi:10.1126/sciadv.adi9127
BibTeX
@article{BrckerhoffPlckelmann2023,
author = {Brückerhoff-Plückelmann, Frank and Bente, Ivonne and Becker, Marlon and Vollmar, Niklas and Farmakidis, Nikolaos and Lomonte, Emma and Lenzini, Francesco and Wright, C David and Bhaskaran, Harish and Salinga, Martin and Risse, Benjamin and Pernice, Wolfram H P},
title = {Event-driven adaptive optical neural network},
journal = {Science Advances},
year = {2023},
volume = {9},
number = {42},
doi = {10.1126/sciadv.adi9127}
}
EUVIMED