Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group
Conferences > ACPC > 2021 > T4A > T4A.244

An Energy-efficient Non-volatile Silicon Photonic Accelerator for Convolutional Neural Networks (NVSP-CNNs)

Zhiyuan Liu, Pengxing Guo, Peng Zhao, Weigang Hou, and Lei Guo

Not Accessible

Your library or personal account may give you access

Abstract

In this paper, we proposed a non-volatile silicon photonic (NVSP) CNN accelerator architecture. Simulation results demonstrate that compared with traditional photonic CNN accelerator DEAP-CNNs, our architecture can reduce the power loss by 29.47%.

© 2021 The Author(s)

PDF Article
More Like This
A Non-Volatile Optical Memory in Silicon Photonics

Jacqueline Geler-Kremer, Felix Eltes, Pascal Stark, Ankita Sharma, Daniele Caimi, Bert Jan Offrein, Jean Fompeyrine, and Stefan Abel
Th4I.2 Optical Fiber Communication Conference (OFC) 2021

Silicon Photonic Filter-based Dot Product Engine for Convolutional Neural Networks

L. De Marinis, M. Sorel, C. Klitis, G. Contestabile, and N. Andriolli
SpM5C.3 Signal Processing in Photonic Communications (SPPCom) 2021

Fourier Optical Convolutional Neural Network Accelerator

Mario Miscuglio, Zibo Hu, Shurui Li, Puneet Gupta, Hamed Dalir, and Volker J. Sorger
SpM5C.2 Signal Processing in Photonic Communications (SPPCom) 2021

View More...
Select as filters
Select Topics Cancel
Top
       
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.
Privacy | Terms of Use