eWB: Event-Based Weight Binarization Algorithm for Spiking Neural Networks
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Dohun | - |
dc.contributor.author | Kim, Guhyun | - |
dc.contributor.author | Hwang, Cheol Seong | - |
dc.contributor.author | Jeong, Doo Seok | - |
dc.date.accessioned | 2021-07-30T04:48:11Z | - |
dc.date.available | 2021-07-30T04:48:11Z | - |
dc.date.created | 2021-07-14 | - |
dc.date.issued | 2021-03 | - |
dc.identifier.issn | 2169-3536 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/1352 | - |
dc.description.abstract | Learning binary weights to minimize the difference between target and actual outputs can be considered as a parameter optimization task within the given constraints, and thus, it belongs to the application domain of the Lagrange multiplier method (LMM). Based on the LMM, we propose a novel event-based weight binarization (eWB) algorithm for spiking neural networks (SNNs) with binary synaptic weights (-1, 1). The algorithm features (i) event-based asymptotic weight binarization using local data only, (ii) full compatibility with event-based end-to-end learning algorithms (e.g., event-driven random backpropagation (eRBP) algorithm), and (iii) the capability to address various constraints (including the binary weight constraint). As a proof of concept, we combine eWB with eRBP (eWB-eRBP) to obtain a single algorithm for learning binary weights to generate correct classifications. Fully connected SNNs were trained using eWB-eRBP and achieved an accuracy of 95.35% on MNIST. To the best of our knowledge, this is the first report on completely binary SNNs trained using an event-based learning algorithm. Given that eRBP with full-precision (32-bit) weights exhibited 97.20% accuracy, the binarization comes at the cost of an accuracy reduction of approximately 1.85%. The python code is available online: https://github.com/galactico7/eWB. | - |
dc.language | 영어 | - |
dc.language.iso | en | - |
dc.publisher | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | - |
dc.title | eWB: Event-Based Weight Binarization Algorithm for Spiking Neural Networks | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Jeong, Doo Seok | - |
dc.identifier.doi | 10.1109/ACCESS.2021.3062405 | - |
dc.identifier.scopusid | 2-s2.0-85101792960 | - |
dc.identifier.wosid | 000628900600001 | - |
dc.identifier.bibliographicCitation | IEEE ACCESS, v.9, pp.38097 - 38106 | - |
dc.relation.isPartOf | IEEE ACCESS | - |
dc.citation.title | IEEE ACCESS | - |
dc.citation.volume | 9 | - |
dc.citation.startPage | 38097 | - |
dc.citation.endPage | 38106 | - |
dc.type.rims | ART | - |
dc.type.docType | Article | - |
dc.description.journalClass | 1 | - |
dc.description.isOpenAccess | Y | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Engineering | - |
dc.relation.journalResearchArea | Telecommunications | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Information Systems | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.relation.journalWebOfScienceCategory | Telecommunications | - |
dc.subject.keywordPlus | Backpropagation | - |
dc.subject.keywordPlus | Lagrange multipliers | - |
dc.subject.keywordPlus | Multiplying circuits | - |
dc.subject.keywordPlus | Asymptotic weight | - |
dc.subject.keywordPlus | Binarization algorithm | - |
dc.subject.keywordPlus | Lagrange multiplier method | - |
dc.subject.keywordPlus | Parameter optimization | - |
dc.subject.keywordPlus | Proof of concept | - |
dc.subject.keywordPlus | Spiking neural networks | - |
dc.subject.keywordPlus | Synaptic weight | - |
dc.subject.keywordPlus | Weight constraints | - |
dc.subject.keywordPlus | Neural networks | - |
dc.subject.keywordAuthor | Neurons | - |
dc.subject.keywordAuthor | Synapses | - |
dc.subject.keywordAuthor | Classification algorithms | - |
dc.subject.keywordAuthor | Approximation algorithms | - |
dc.subject.keywordAuthor | Linear programming | - |
dc.subject.keywordAuthor | Training | - |
dc.subject.keywordAuthor | System-on-chip | - |
dc.subject.keywordAuthor | Event-based weight binarization | - |
dc.subject.keywordAuthor | event-driven learning algorithm | - |
dc.subject.keywordAuthor | Lagrange multiplier method | - |
dc.subject.keywordAuthor | spiking neural networks | - |
dc.identifier.url | https://ieeexplore.ieee.org/document/9363894 | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
222, Wangsimni-ro, Seongdong-gu, Seoul, 04763, Korea+82-2-2220-1365
COPYRIGHT © 2021 HANYANG UNIVERSITY.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.