Taha Shahroodi, Raphael Cardoso, Mahdi Zahedi, Stephan Wong, A. Bosio, Ian O’Connor, S. Hamdioui
{"title":"利用光学相变材料的光速二元神经网络","authors":"Taha Shahroodi, Raphael Cardoso, Mahdi Zahedi, Stephan Wong, A. Bosio, Ian O’Connor, S. Hamdioui","doi":"10.23919/DATE56975.2023.10137229","DOIUrl":null,"url":null,"abstract":"This paper investigates the potential of a compute-in-memory core based on optical Phase Change Materials (oPCMs) to speed up and reduce the energy consumption of the Matrix-Matrix-Multiplication operation. The paper also proposes a new data mapping for Binary Neural Networks (BNNs) tailored for our oPCM core. The preliminary results show a significant latency improvement irrespective of the evaluated network structure and size. The improvement varies from network to network and goes up to ~1053x.","PeriodicalId":340349,"journal":{"name":"2023 Design, Automation & Test in Europe Conference & Exhibition (DATE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Lightspeed Binary Neural Networks using Optical Phase-Change Materials\",\"authors\":\"Taha Shahroodi, Raphael Cardoso, Mahdi Zahedi, Stephan Wong, A. Bosio, Ian O’Connor, S. Hamdioui\",\"doi\":\"10.23919/DATE56975.2023.10137229\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper investigates the potential of a compute-in-memory core based on optical Phase Change Materials (oPCMs) to speed up and reduce the energy consumption of the Matrix-Matrix-Multiplication operation. The paper also proposes a new data mapping for Binary Neural Networks (BNNs) tailored for our oPCM core. The preliminary results show a significant latency improvement irrespective of the evaluated network structure and size. The improvement varies from network to network and goes up to ~1053x.\",\"PeriodicalId\":340349,\"journal\":{\"name\":\"2023 Design, Automation & Test in Europe Conference & Exhibition (DATE)\",\"volume\":\"30 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-04-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2023 Design, Automation & Test in Europe Conference & Exhibition (DATE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.23919/DATE56975.2023.10137229\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 Design, Automation & Test in Europe Conference & Exhibition (DATE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/DATE56975.2023.10137229","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Lightspeed Binary Neural Networks using Optical Phase-Change Materials
This paper investigates the potential of a compute-in-memory core based on optical Phase Change Materials (oPCMs) to speed up and reduce the energy consumption of the Matrix-Matrix-Multiplication operation. The paper also proposes a new data mapping for Binary Neural Networks (BNNs) tailored for our oPCM core. The preliminary results show a significant latency improvement irrespective of the evaluated network structure and size. The improvement varies from network to network and goes up to ~1053x.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
