Summary
International Symposium on Nonlinear Theory and Its Applications
2022
Session Number:B1L-C
Session:
Number:B1L-C-01
Photonic Neural Field Dynamics and its Application to Reservoir Computing
Kohei Arai , Tomoya Yamaguchi , Tomoaki Niiyama , Satoshi Sunada,
pp.219-220
Publication Date:12/12/2022
Online ISSN:2188-5079
DOI:10.34385/proc.71.B1L-C-01
PDF download (726.8KB)
Summary:
Photonic reservoir computing hardware has been expected as a novel photonic hardware enabling high- speed and high-efficiency information processing, but it remains challenges in terms of the network scalability. In this study, we introduce a concept of photonic neural field, which corresponds to a spatially continuous neural network, and show the implementation of the photonic neural field on a silicon chip. We demonstrate that the photonic neural field on a silicon chip is capable of high-density and large- scale reservoir network processing for benchmark tasks, including chaotic time series prediction and image classification.