Show simple item record

dc.contributor.authorSchmuker, Michael
dc.contributor.authorPfeil, Thomas
dc.contributor.authorNawrot, Martin Paul
dc.date.accessioned2017-09-13T15:57:56Z
dc.date.available2017-09-13T15:57:56Z
dc.date.issued2014-02-11
dc.identifier.citationSchmuker , M , Pfeil , T & Nawrot , M P 2014 , ' A neuromorphic network for generic multivariate data classification ' Proceedings of the National Academy of Sciences of the United States of America , vol. 111 , no. 6 , pp. 2081-2086 . https://doi.org/10.1073/pnas.1303053111
dc.identifier.issn0027-8424
dc.identifier.otherPURE: 10467425
dc.identifier.otherPURE UUID: f3c531b8-aace-4cdf-aac7-a61e1faafe0b
dc.identifier.otherPubMed: 24469794
dc.identifier.otherScopus: 84893861508
dc.identifier.urihttp://hdl.handle.net/2299/19378
dc.descriptionMichael Schmuker, Thomas Pfeil, and Martin Paul Nawrot, ‘A neuromorphic network for generic multivariate data classification’, Proceedings of the National Academy of Sciences of the United States of America, Vol. 111 (6): 2081-2081, February 2014, doi: http://dx.doi.org/10.1073/pnas.1303053111.
dc.description.abstractComputational neuroscience has uncovered a number of computational principles used by nervous systems. At the same time, neuromorphic hardware has matured to a state where fast silicon implementations of complex neural networks have become feasible. En route to future technical applications of neuromorphic computing the current challenge lies in the identification and implementation of functional brain algorithms. Taking inspiration from the olfactory system of insects, we constructed a spiking neural network for the classification of multivariate data, a common problem in signal and data analysis. In this model, real-valued multivariate data are converted into spike trains using "virtual receptors" (VRs). Their output is processed by lateral inhibition and drives a winner-take-all circuit that supports supervised learning. VRs are conveniently implemented in software, whereas the lateral inhibition and classification stages run on accelerated neuromorphic hardware. When trained and tested on real-world datasets, we find that the classification performance is on par with a naïve Bayes classifier. An analysis of the network dynamics shows that stable decisions in output neuron populations are reached within less than 100 ms of biological time, matching the time-to-decision reported for the insect nervous system. Through leveraging a population code, the network tolerates the variability of neuronal transfer functions and trial-to-trial variation that is inevitably present on the hardware system. Our work provides a proof of principle for the successful implementation of a functional spiking neural network on a configurable neuromorphic hardware system that can readily be applied to real-world computing problems.en
dc.format.extent6
dc.language.isoeng
dc.relation.ispartofProceedings of the National Academy of Sciences of the United States of America
dc.rights/dk/atira/pure/core/openaccesspermission/open
dc.subjectAlgorithms
dc.subjectBayes Theorem
dc.subjectBrain
dc.subjectLearning
dc.subjectMultivariate Analysis
dc.subjectNeural Networks (Computer)
dc.titleA neuromorphic network for generic multivariate data classificationen
dc.contributor.institutionSchool of Computer Science
dc.description.statusPeer reviewed
dc.relation.schoolSchool of Computer Science
dc.description.versiontypeFinal Published version
rioxxterms.versionVoR
rioxxterms.versionofrecordhttps://doi.org/10.1073/pnas.1303053111
rioxxterms.typeJournal Article/Review
herts.preservation.rarelyaccessedtrue
herts.rights.accesstypeopenAccess


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record