Silicon microring synapses enable photonic deep learning beyond 9-bit precision

Deep neural networks (DNNs) consist of layers of neurons interconnected by synaptic weights. A high bit-precision in weights is generally required to guarantee high accuracy in many applications. Minimizing error accumulation between layers is also essential when building large-scale networks. Recen...

Full description

Saved in:
Bibliographic Details
Published inOptica Vol. 9; no. 5; p. 579
Main Authors Zhang, Weipeng, Huang, Chaoran, Peng, Hsuan-Tung, Bilodeau, Simon, Jha, Aashu, Blow, Eric, de Lima, Thomas Ferreira, Shastri, Bhavin J., Prucnal, Paul
Format Journal Article
LanguageEnglish
Published 20.05.2022
Online AccessGet full text

Cover

More Information
Summary:Deep neural networks (DNNs) consist of layers of neurons interconnected by synaptic weights. A high bit-precision in weights is generally required to guarantee high accuracy in many applications. Minimizing error accumulation between layers is also essential when building large-scale networks. Recent demonstrations of photonic neural networks are limited in bit-precision due to cross talk and the high sensitivity of optical components (e.g., resonators). Here, we experimentally demonstrate a record-high precision of 9 bits with a dithering control scheme for photonic synapses. We then numerically simulated the impact with increased synaptic precision on a wireless signal classification application. This work could help realize the potential of photonic neural networks for many practical, real-world tasks.
ISSN:2334-2536
2334-2536
DOI:10.1364/OPTICA.446100