A Hybrid Optical-Electrical Analog Deep Learning Accelerator Using Incoherent Optical Signals

Mingdai Yang, Mohammad Reza Jokar, Junyi Qiu, Qiuwen Lou, Yuming Liu, Aditi Udupa, Frederic T. Chong, John M. Dallesasse, Milton Feng, Lynford L. Goddard, X. Sharon Hu, Yanjing Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a hybrid optical-electrical analog deep learning (DL) accelerator, the first work to use incoherent optical signals for DL workloads. Incoherent optical designs are more attractive than coherent ones as the former can be more easily realized in practice. However, a significant challenge in analog DL accelerators, where multiply-Accumulate operations are dominant, is that there is no known solution to perform accumulation using incoherent optical signals. We overcome this challenge by devising a hybrid approach: Accumulation is done in the electrical domain, while multiplication is performed in the optical domain. The key technology enabler of our design is the transistor laser, which performs electrical-To-optical and optical-To-electrical conversions efficiently to tightly integrate electrical and optical devices into compact circuits. As such, our design fully realizes the ultra high-speed and high-energy-efficiency advantages of analog and optical computing. Our evaluation results using the MNIST benchmark show that our design achieves 2214× and 65× improvements in latency and energy, respectively, compared to a state-of-The-Art memristor-based analog design.

Original languageEnglish (US)
Title of host publicationGLSVLSI 2021 - Proceedings of the 2021 Great Lakes Symposium on VLSI
PublisherAssociation for Computing Machinery
Pages271-276
Number of pages6
ISBN (Electronic)9781450383936
DOIs
StatePublished - Jun 22 2021
Event31st Great Lakes Symposium on VLSI, GLSVLSI 2021 - Virtual, Online, United States
Duration: Jun 22 2021Jun 25 2021

Publication series

NameProceedings of the ACM Great Lakes Symposium on VLSI, GLSVLSI

Conference

Conference31st Great Lakes Symposium on VLSI, GLSVLSI 2021
Country/TerritoryUnited States
CityVirtual, Online
Period6/22/216/25/21

Keywords

  • deep learning accelerator
  • optical computing

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint

Dive into the research topics of 'A Hybrid Optical-Electrical Analog Deep Learning Accelerator Using Incoherent Optical Signals'. Together they form a unique fingerprint.

Cite this