Reinforcement Learning for Photonic Component Design

Donald Witt, Jeff Young, Lukas Chrostowski
7/14/2023
physics.opticscs.LGphysics.app-ph

Official code: Donald-Witt/Fab-in-the-loop (6 stars, GPL-3.0, last commit 2023-10-06).

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.

View on arXivView PDF

Code Implementations(1)

Donald-Witt/Fab-in-the-loopOfficial
GPL-3.0

Code for fab-in-the-loop reinforcement learning

60PythonApr 3, 20233 years agoGPL-3.0

Similar Papers

Multi-Agent Reinforcement Learning for Inverse Design in Photonic Integrated Circuits

83% match2025

Inverse design of photonic surfaces on Inconel via multi-fidelity machine learning ensemble framework and high throughput femtosecond laser processing

79% match2024

MAPS: Multi-Fidelity AI-Augmented Photonic Simulation and Inverse Design Infrastructure

79% match2025

Democratizing Electronic-Photonic AI Systems: An Open-Source AI-Infused Cross-Layer Co-Design and Design Automation Toolflow

78% match2025

Multi-objective Bayesian Optimization with Human-in-the-Loop for Flexible Neuromorphic Electronics Fabrication

78% match2025

Artificial Neural Networks for Photonic Applications: From Algorithms to Implementation

78% match2024

Cite this paper

@article{witt2023reinforcement,
  title  = {Reinforcement Learning for Photonic Component Design},
  author = {Donald Witt and Jeff Young and Lukas Chrostowski},
  year   = {2023},
  eprint = {2307.11075},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2307.11075v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2307.11075.svg)](https://codeofpaper.com/paper/2307.11075)

Discussion