Photonics Inverse Design: Pairing Deep Neural Networks With Evolutionary Algorithms

• Published in: IEEE journal of selected topics in quantum electronics Vol. 26; no. 1; pp. 1 - 8
• Main Author: Hegde, Ravi S.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: artificial intelligence; Artificial neural networks; deep learning; Evolutionary algorithms; Evolutionary computation; evolutionary optimization; Genetic algorithms; Inverse design; Iterative methods; Neural networks; Optical system design; optics and lens design; Optimization; Optimization techniques; Photonics; Python; Sociology; Statistics; thin films; Training; Workflow
• ISSN: 1077-260X; 1558-4542
• DOI: 10.1109/JSTQE.2019.2933796

Deep Neural Networks (DNN) have shown early promise for inverse design with their ability to arrive at working designs much faster than conventional optimization techniques. Current approaches, however, require complicated workflows involving training more than one DNN to address the problem of non-uniqueness in the inversion and the emphasis on speed has overshadowed the far more important consideration of solution optimality. We propose and demonstrate a simplified workflow that pairs forward-model DNN with evolutionary algorithms which are widely used for inverse gg design. Our evolutionary search in forward-model space is global and exploits the massive parallelism of modern GPUs for a speedy inversion. We propose a hybrid approach where the DNN is used only for preselection and initialization that is more effective at optimization than a standalone DNN and performs nearly as well as a vanilla evolutionary search with a significantly reduced function evaluation budget. We finally show the utility of an iterative procedure for building the training dataset which further boosts the effectiveness of this approach.