Reservoir Computing Using Autonomous Boolean Networks Realized on Field-Programmable Gate Arrays

dc.contributor.authorApostel, Stefan
dc.contributor.authorHaynes, Nicholas D.
dc.contributor.authorSchöll, Eckehard
dc.contributor.authorD’Huys, Otti
dc.contributor.authorGauthier, Daniel J.
dc.date.accessioned2022-01-27T20:18:04Z
dc.date.available2022-01-27T20:18:04Z
dc.date.issued2021-08-06
dc.description.abstractIn this chapter, we consider realizing a reservoir computer on an electronic chip that allows for many tens of network nodes whose connection topology can be quickly reconfigured. The reservoir computer displays analog-like behavior and has the potential to perform computations beyond that of a classic Turning machine. In detail, we present our preliminary results of using a physical reservoir computer for performing the task of identifying written digits. The reservoir is realized on a commercially available electronic device known as a field-programmable gate array on which we create an autonomous Boolean network for information processing. Even though the network nodes are Boolean logic elements, they display analog behavior because there is no master clock that controls the nodes. In addition, the electronic signals related to the written-digit images are injected into the reservoir at high speed, leading to the possibility of full-image classification on the nanosecond time scale. We explore the dynamics of the autonomous Boolean networks in response to injected signals and, based on these results, investigate the performance of the reservoir computer on the written-digit task. For a wide range of reservoir structures, we obtain a typical performance of ∼ 90% for correctly identifying a written digit, which exceeds that obtained by a linear classifier. This work paves the way for achieving low-power, high-speed reservoir computing on readily available field-programmable gate arrays, which are well matched to existing computing infrastructure.en
dc.description.sponsorshipDFG, 163436311, SFB 910: Kontrolle selbstorganisierender nichtlinearer Systeme: Theoretische Methoden und Anwendungskonzepteen
dc.description.sponsorshipEC/H2020/713694/EU/International Mobility and Training in Photonics Programme/MULTIPLYen
dc.identifier.eissn1619-7127
dc.identifier.isbn978-981-13-1687-6
dc.identifier.isbn978-981-13-1686-9
dc.identifier.urihttps://depositonce.tu-berlin.de/handle/11303/16211
dc.identifier.urihttp://dx.doi.org/10.14279/depositonce-14985
dc.language.isoenen
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subject.ddc004 Datenverarbeitung; Informatikde
dc.subject.ddc530 Physikde
dc.subject.othermachine learningen
dc.subject.otherdata classificationen
dc.subject.otherreservoir computingen
dc.subject.otherfield-programmable gate arrayen
dc.subject.otherrecurrent networken
dc.subject.otherMNISTen
dc.subject.otherphase transitionen
dc.subject.othernetwork dynamicsen
dc.subject.otherautonomous Boolean networksen
dc.titleReservoir Computing Using Autonomous Boolean Networks Realized on Field-Programmable Gate Arraysen
dc.typeBook Parten
dc.type.versionacceptedVersionen
dcterms.bibliographicCitation.booktitleReservoir Computing: Theory, Physical Implementations, and Applicationsen
dcterms.bibliographicCitation.doi10.1007/978-981-13-1687-6_11en
dcterms.bibliographicCitation.editorNakajima, Kohei
dcterms.bibliographicCitation.editorFischer, Ingo
dcterms.bibliographicCitation.originalpublishernameSpringeren
dcterms.bibliographicCitation.originalpublisherplaceSingaporeen
dcterms.bibliographicCitation.pageend271en
dcterms.bibliographicCitation.pagestart239en
tub.accessrights.dnbdomain*
tub.affiliationFak. 2 Mathematik und Naturwissenschaften::Inst. Theoretische Physik::FG Nichtlineare Dynamik und Kontrollede
tub.affiliation.facultyFak. 2 Mathematik und Naturwissenschaftende
tub.affiliation.groupFG Nichtlineare Dynamik und Kontrollede
tub.affiliation.instituteInst. Theoretische Physikde
tub.publisher.universityorinstitutionTechnische Universität Berlinen

Files

Original bundle
Now showing 1 - 1 of 1
Loading…
Thumbnail Image
Name:
Apostel_etal_2021.pdf
Size:
2.22 MB
Format:
Adobe Portable Document Format
Description:

Collections