Photonic delay-based reservoir computers as deep neural network preprocessors

Research output: Chapter in Book/Report/Conference proceedingConference paper

Abstract

Photonic reservoir computing (RC) has been effectively used for solving various complex problems. Such a reservoir consists of a network of randomly, untrained connected nodes. Doing RC in the photonics domain offers the advantage of high-speed performance, low-energy consumption and the possibility of high inherent parallelism. We propose and numerically investigate to use the output of such a reservoir to preprocess the input data before this data is send to a DNN. The main idea here is to use such a photonic reservoir to transform the input data into a higher dimensional state-space, which could allow the DNN to process the data with increased performance. Based on numerical simulations of delay-based reservoirs using a single-mode semiconductor laser, we show that using such a preprocessing reservoir results in an improved performance of DNNs, and that we do not need to carefully fine-tune the parameters of the preprocessing reservoir.
Original languageEnglish
Title of host publicationConference on Lasers and Electro-Optics/Europe (CLEO/Europe 2023)
Pages8-8
Number of pages1
Publication statusPublished - 30 Jun 2023
EventCLEO/Europe-EQEC 2023 - Munich, Germany
Duration: 26 Jun 202330 Jun 2023

Conference

ConferenceCLEO/Europe-EQEC 2023
Country/TerritoryGermany
CityMunich
Period26/06/2330/06/23

Fingerprint

Dive into the research topics of 'Photonic delay-based reservoir computers as deep neural network preprocessors'. Together they form a unique fingerprint.

Cite this