DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kwon, Eunji | - |
dc.contributor.author | Kang, Yesung | - |
dc.contributor.author | Kang, Seokhyeong | - |
dc.date.accessioned | 2021-06-01T11:51:38Z | - |
dc.date.available | 2021-06-01T11:51:38Z | - |
dc.date.created | 2021-03-10 | - |
dc.date.issued | 2019-10-08 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/106390 | - |
dc.description.abstract | Convolutional neural networks (CNNs) are computationally intensive, and deep learning hardware should be implemented energy-efficiently for embedded systems or battery-constrained systems. In this paper, we propose an outlier-Aware time-multiplexing MAC. We exploit a CNN feature maps' characteristic of being able to express most of the data in a low bit-width except a few large values, which we call 'outliers' Our outlier-Aware time-multiplexing MAC has improved the energy efficiency by up to 21.1% compared to conventional MACs. | - |
dc.language | English | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.relation.isPartOf | 16th International System-on-Chip Design Conference, ISOCC 2019 | - |
dc.relation.isPartOf | Proceedings - 2019 International SoC Design Conference, ISOCC 2019 | - |
dc.title | Outlier-Aware Time-multiplexing MAC for Higher Energy-Efficiency on CNNs | - |
dc.type | Conference | - |
dc.type.rims | CONF | - |
dc.identifier.bibliographicCitation | 16th International System-on-Chip Design Conference, ISOCC 2019, pp.119 - 120 | - |
dc.citation.conferenceDate | 2019-10-06 | - |
dc.citation.conferencePlace | KO | - |
dc.citation.endPage | 120 | - |
dc.citation.startPage | 119 | - |
dc.citation.title | 16th International System-on-Chip Design Conference, ISOCC 2019 | - |
dc.contributor.affiliatedAuthor | Kang, Seokhyeong | - |
dc.identifier.scopusid | 2-s2.0-85082989764 | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.