[CThA13B] Photonic Computing I

[CThA13B-02 (Invited)] Silicon Photonics for Training Deep Neural Networks

*Bhavin J. Shastri^1,4, Matthew J. Filipovich¹, Zhimu Guo¹, Paul R Prucnal⁴, Sudip Shekhar², Volker J. Sorger³ (1. Queen's University (Canada), 2. University of British Columbia (Canada), 3. George Washington University (United States of America), 4. Princeton University (United States of America))

[Presentation Style] Online

Analog photonic networks as deep learning hardware accelerators are trained on standard digital electronics. We propose an on-chip training of neural networks enabled by a silicon photonic architecture for parallel, efficient, and fast data operations.