Reinforcement Learning for Photonic Component Design
This addresses the challenge of optimizing photonic component performance under real-world fabrication constraints for applications in photonics and nanotechnology, representing a strong specific gain rather than a broad paradigm shift.
The researchers tackled the problem of designing nano-photonic components by accounting for fabrication imperfections, achieving an improvement in insertion loss from 8.8 to 3.24 dB and enabling designs with a 150 nm bandwidth and less than 10.2 dB loss.
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.