论文信息 - Shape design of a reflecting surface using Bayesian Optimization

Shape design of a reflecting surface using Bayesian Optimization

Optimizing free-from shapes of photonic nanostructures is a high-dimensional problem. We present a Bayesian optimization algorithm with a hyper-parameter learning routine. We apply the algorithm for optimizing the shape of a reflecting meta-surface. The algorithm exhibits efficient performance.

[1] Ilya Fushman,et al. Genetic optimization of photonic bandgap structures. , 2007, Optics express.

[2] Carsten Rockstuhl,et al. Global optimization of complex optical structures using Bayesian optimization based on Gaussian processes , 2017, Other Conferences.

[3] Yifang Chen,et al. Nanofabrication by electron beam lithography and its applications , 2015 .

[4] Luca Dal Negro,et al. Particle-swarm optimization of broadband nanoplasmonic arrays. , 2010, Optics letters.

[5] Frank Schmidt,et al. Adaptive finite element method for simulation of optical nano structures , 2007, 0711.2149.

[6] B. Shen,et al. An integrated-nanophotonics polarization beamsplitter with 2.4 × 2.4 μm2 footprint , 2015, Nature Photonics.

[7] J. Fischer,et al. Three‐dimensional optical laser lithography beyond the diffraction limit , 2013 .

[8] Hod Lipson,et al. Two-dimensional photonic crystals designed by evolutionary algorithms , 2005 .