Design Automation of Photonic Resonator Weights

TL;DR

This paper reviews the challenges and solutions for designing robust photonic resonator weight banks for neuromorphic AI, focusing on automation, control, and compensation techniques to enable practical deployment.

Contribution

It introduces automated design and control methodologies to create high-precision resonator weight banks resilient to fabrication and environmental variations.

Findings

Mathematical modeling of variations and disturbances

Exploitation of resonator physics for weighting and summing signals

Road map for practical deployment of resonator weight banks

Abstract

Neuromorphic photonic processors based on resonator weight banks are an emerging candidate technology for enabling modern artificial intelligence (AI) in high speed, analog systems. These purpose-built analog devices implement vector multiplications with the physics of resonator devices, offering efficiency, latency, and throughput advantages over equivalent electronic circuits. Along with these advantages, however, often comes the difficult challenges of compensation for fabrication variations and environmental disturbances. In this paper we review sources of variation and disturbances from our experiments, as well as mathematically define quantities that model them. Then, we introduce how the physics of resonators can be exploited to weight and sum multiwavelength signals. Finally, we outline automated design and control methodologies necessary to create practical, manufacturable, and high accuracy/precision resonator weight banks that can withstand operating conditions in the field. This represents a road map for unlocking the potential of resonator weight banks in practical deployment scenarios.