Group Refractive Index Via Auto-differentiation And Neural Networks

Interest in photonics research is growing, and accelerated design and modelling tools are needed to improve the state-of-the-art components and systems. Brute-force numerical methods are often used to explore the limited parameter space, but this is inefficient. Deep learning models can be used as a replacement for traditional photonic models, and automatic design techniques can be used to further improve the design process.