Reinforcement Learning for Photonic Component Design

TL;DR

This paper introduces a reinforcement learning algorithm that incorporates fabrication imperfections for designing nano-photonic components, demonstrating significant improvements in device performance and bandwidth coverage.

Contribution

The paper presents a novel fab-in-the-loop reinforcement learning method specifically tailored for nano-photonic component design, accounting for manufacturing imperfections.

Findings

Reduced insertion loss from 8.8 to 3.24 dB

Achieved 150 nm bandwidth with less than 10.2 dB loss

Demonstrated effectiveness on silicon photonic crystal grating couplers

Abstract

We present a new fab-in-the-loop reinforcement learning algorithm for the design of nano-photonic components that accounts for the imperfections present in nanofabrication processes. As a demonstration of the potential of this technique, we apply it to the design of photonic crystal grating couplers fabricated on an air clad 220 nm silicon on insulator single etch platform. This fab-in-the-loop algorithm improves the insertion loss from 8.8 to 3.24 dB. The widest bandwidth designs produced using our fab-in-the-loop algorithm can cover a 150 nm bandwidth with less than 10.2 dB of loss at their lowest point.