Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Hwang, Sangwoo | - |
dc.contributor.author | Lee, Junghyup | - |
dc.contributor.author | Kung, Jaeha | - |
dc.date.accessioned | 2023-12-26T18:44:19Z | - |
dc.date.available | 2023-12-26T18:44:19Z | - |
dc.date.created | 2021-07-16 | - |
dc.date.issued | 2021-05-24 | - |
dc.identifier.isbn | 9781728192017 | - |
dc.identifier.issn | 2158-1525 | - |
dc.identifier.uri | http://hdl.handle.net/20.500.11750/46930 | - |
dc.description.abstract | In this paper, we present a novel approach in developing input-to-neuron interlinks to achieve better accuracy in spike-based liquid state machines. An energy-efficient Spiking Neural Network suffer from lower accuracy in image classification compared to deep learning models. The previous LSM models randomly connect input neurons to excitatory neurons in a liquid. This limits the expressive power of a liquid model as large portion of excitatory neurons become inactive which never fire. To overcome this limitation, we propose an adaptive interlink development method which achieves 3.2% higher classification accuracy than the static LSM model of 3,200 neurons. Also, our hardware implementation on FPGA improves performance by 3.16∼4.99× or 1.47∼3.95× over CPU/GPU. © 2021 IEEE | - |
dc.language | English | - |
dc.publisher | IEEE Circuits and Systems Society | - |
dc.title | Adaptive Input-to-Neuron Interlink Development in Training of Spike-Based Liquid State Machines | - |
dc.type | Conference Paper | - |
dc.identifier.doi | 10.1109/ISCAS51556.2021.9401085 | - |
dc.identifier.scopusid | 2-s2.0-85109004462 | - |
dc.identifier.bibliographicCitation | IEEE International Symposium on Circuits and Systems (ISCAS 2021), pp.382 - 386 | - |
dc.identifier.url | https://www.proceedings.com/content/059/059501webtoc.pdf | - |
dc.citation.conferencePlace | KO | - |
dc.citation.conferencePlace | 대구 | - |
dc.citation.endPage | 386 | - |
dc.citation.startPage | 382 | - |
dc.citation.title | IEEE International Symposium on Circuits and Systems (ISCAS 2021) | - |
There are no files associated with this item.