In-memory photonic dot-product engine with electrically programmable weight banks
Abstract
Electronically reprogrammable photonic circuits based on phase-change chalcogenides present an avenue to resolve the von-Neumann bottleneck; however, implementation of such hybrid photonic-electronic processing has not achieved computational success. Here, we achieve this milestone by demonstrating an in-memory photonic-electronic dot-product engine, one that decouples electronic programming of phase-change materials (PCMs) and photonic computation. Specifically, we develop non-volatile electronically reprogrammable PCM memory cells with a record-high 4-bit weight encoding, the lowest energy consumption per unit modulation depth (1.7 nJ per dB) for Erase operation (crystallization), and a high switching contrast (158.5%) using non-resonant silicon-on-insulator waveguide microheater devices. This enables us to perform parallel multiplications for image processing with a superior contrast-to-noise ratio (greater than 87.36) that leads to an enhanced computing accuracy (standard deviation less than 0.007). An in-memory hybrid computing system is developed in hardware for convolutional processing for recognizing images from the MNIST database with inferencing accuracies of 86% and 87%.
Cite
@article{arxiv.2304.14302,
title = {In-memory photonic dot-product engine with electrically programmable weight banks},
author = {Wen Zhou and Bowei Dong and Nikolaos Farmakidis and Xuan Li and Nathan Youngblood and Kairan Huang and Yuhan He and C. David Wright and Wolfram H. P. Pernice and Harish Bhaskaran},
journal= {arXiv preprint arXiv:2304.14302},
year = {2023}
}