Silicon photonic architecture for training deep neural networks with direct feedback alignment
Abstract
There has been growing interest in using photonic processors for performing neural network inference operations; however, these networks are currently trained using standard digital electronics. Here, we propose on-chip training of neural networks enabled by a CMOS-compatible silicon photonic architecture to harness the potential for massively parallel, efficient, and fast data operations. Our scheme employs the direct feedback alignment training algorithm, which trains neural networks using error feedback rather than error backpropagation, and can operate at speeds of trillions of multiply–accumulate (MAC) operations per second while consuming less than one picojoule per MAC operation. The photonic architecture exploits parallelized matrix–vector multiplications using arrays of microring resonators for processing multi-channel analog signals along single waveguide buses to calculate the gradient vector for each neural network layerin situ. We also experimentally demonstrate training deep neural networks with the MNIST dataset using on-chip MAC operation results. Our approach for efficient, ultra-fast neural network training showcases photonics as a promising platform for executing artificial intelligence applications.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Nov 23, 2022
- Source ID
- 10.1364/optica.475493
Entities
People
- Bhavin J. Shastri
- Bicky A. Marquez
- Hugh D. Morison
- Matthew J. Filipovich
- Mohammed Al-qadasi
- Paul Prucnal
- Sudip Shekhar
- Volker Sorger
- Zhimu Guo
Organizations
- Air Force Office of Scientific Research
- Canada Foundation for Innovation
- George Washington University
- Natural Sciences and Engineering Research Council
- Princeton University
- Queen's University
- State Research Center of Virology and Biotechnology VECTOR
- University of British Columbia