Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Physical Limitation Aware Quantization Model for Photonic Convolutional Neural Network

Not Accessible

Your library or personal account may give you access

Abstract

We propose a physical limitation aware clustered quantization model by considering precision device limitation in the training process of integrated photonic CNN, showing 98.5% recognition accuracy for MNIST handwirtten recognition with 2-bit precision.

© 2021 The Author(s)

PDF Article
More Like This
Deep nonlinear optical neural networks using physics-aware training

Logan G. Wright, Tatsuhiro Onodera, Martin M. Stein, Tianyu Wang, Darren T. Schachter, Zoey Hu, and Peter L. McMahon
FF1A.4 CLEO: QELS_Fundamental Science (CLEO:FS) 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 Optic Convolutional Neural Network

Mario Miscuglio, Zibo Hu, Shurui Li, Puneet Gupta, Hamed Dalir, and Volker J. Sorger
ITh5D.6 Imaging Systems and Applications (IS) 2021

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All Rights Reserved