Reservoir Computing Using Autonomous Boolean Networks Realized on Field-Programmable Gate Arrays
dc.contributor.author | Apostel, Stefan | |
dc.contributor.author | Haynes, Nicholas D. | |
dc.contributor.author | Schöll, Eckehard | |
dc.contributor.author | D’Huys, Otti | |
dc.contributor.author | Gauthier, Daniel J. | |
dc.date.accessioned | 2022-01-27T20:18:04Z | |
dc.date.available | 2022-01-27T20:18:04Z | |
dc.date.issued | 2021-08-06 | |
dc.description.abstract | In 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.sponsorship | DFG, 163436311, SFB 910: Kontrolle selbstorganisierender nichtlinearer Systeme: Theoretische Methoden und Anwendungskonzepte | en |
dc.description.sponsorship | EC/H2020/713694/EU/International Mobility and Training in Photonics Programme/MULTIPLY | en |
dc.identifier.eissn | 1619-7127 | |
dc.identifier.isbn | 978-981-13-1687-6 | |
dc.identifier.isbn | 978-981-13-1686-9 | |
dc.identifier.uri | https://depositonce.tu-berlin.de/handle/11303/16211 | |
dc.identifier.uri | http://dx.doi.org/10.14279/depositonce-14985 | |
dc.language.iso | en | en |
dc.rights.uri | http://rightsstatements.org/vocab/InC/1.0/ | en |
dc.subject.ddc | 004 Datenverarbeitung; Informatik | de |
dc.subject.ddc | 530 Physik | de |
dc.subject.other | machine learning | en |
dc.subject.other | data classification | en |
dc.subject.other | reservoir computing | en |
dc.subject.other | field-programmable gate array | en |
dc.subject.other | recurrent network | en |
dc.subject.other | MNIST | en |
dc.subject.other | phase transition | en |
dc.subject.other | network dynamics | en |
dc.subject.other | autonomous Boolean networks | en |
dc.title | Reservoir Computing Using Autonomous Boolean Networks Realized on Field-Programmable Gate Arrays | en |
dc.type | Book Part | en |
dc.type.version | acceptedVersion | en |
dcterms.bibliographicCitation.booktitle | Reservoir Computing: Theory, Physical Implementations, and Applications | en |
dcterms.bibliographicCitation.doi | 10.1007/978-981-13-1687-6_11 | en |
dcterms.bibliographicCitation.editor | Nakajima, Kohei | |
dcterms.bibliographicCitation.editor | Fischer, Ingo | |
dcterms.bibliographicCitation.originalpublishername | Springer | en |
dcterms.bibliographicCitation.originalpublisherplace | Singapore | en |
dcterms.bibliographicCitation.pageend | 271 | en |
dcterms.bibliographicCitation.pagestart | 239 | en |
tub.accessrights.dnb | domain | * |
tub.affiliation | Fak. 2 Mathematik und Naturwissenschaften::Inst. Theoretische Physik::FG Nichtlineare Dynamik und Kontrolle | de |
tub.affiliation.faculty | Fak. 2 Mathematik und Naturwissenschaften | de |
tub.affiliation.group | FG Nichtlineare Dynamik und Kontrolle | de |
tub.affiliation.institute | Inst. Theoretische Physik | de |
tub.publisher.universityorinstitution | Technische Universität Berlin | en |
Files
Original bundle
1 - 1 of 1
Loading…
- Name:
- Apostel_etal_2021.pdf
- Size:
- 2.22 MB
- Format:
- Adobe Portable Document Format
- Description: