Optimizing High-Dimensional Physics Simulations via Composite Bayesian Optimization

Physical simulation-based optimization is a common task in science and engineering. Many such simulations produce imageor tensor-based outputs where the desired objective is a function of those outputs, and optimization is performed over a high-dimensional parameter space. We develop a Bayesian optimization method leveraging tensor-based Gaussian process surrogates and trust region Bayesian optimization to effectively model the image outputs and to efficiently optimize these types of simulations, including a radio-frequency tower configuration problem and an optical design problem.

[1]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[2]  Peter I. Frazier,et al.  Bayesian optimization for materials design , 2015, 1506.01349.

[3]  J Duris,et al.  Bayesian Optimization of a Free-Electron Laser. , 2020, Physical review letters.

[4]  Eytan Bakshy,et al.  Multi-Objective Bayesian Optimization over High-Dimensional Search Spaces , 2021, ArXiv.

[5]  Robert W. Heath,et al.  Optimizing Coverage and Capacity in Cellular Networks using Machine Learning , 2021, ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[6]  Timothy C. Coburn,et al.  Geostatistics for Natural Resources Evaluation , 2000, Technometrics.

[7]  Matthias Poloczek,et al.  Scalable Global Optimization via Local Bayesian Optimization , 2019, NeurIPS.

[8]  Stefan M. Wild,et al.  Bayesian Calibration and Uncertainty Analysis for Computationally Expensive Models Using Optimization and Radial Basis Function Approximation , 2008 .

[9]  Wei Chen,et al.  Bayesian Optimization for Materials Design with Mixed Quantitative and Qualitative Variables , 2019, Scientific Reports.

[10]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[11]  Andrew Gordon Wilson,et al.  Bayesian Optimization with High-Dimensional Outputs , 2021, NeurIPS.

[12]  Matthias Poloczek,et al.  Scalable Constrained Bayesian Optimization , 2020, AISTATS.

[13]  Daniel R. Jiang,et al.  BoTorch: A Framework for Efficient Monte-Carlo Bayesian Optimization , 2020, NeurIPS.

[14]  Shandian Zhe,et al.  Scalable High-Order Gaussian Process Regression , 2019, AISTATS.

[15]  D. Birchall,et al.  Computational Fluid Dynamics , 2020, Radial Flow Turbocompressors.

[16]  D. O'shea,et al.  Diffractive Optics: Design, Fabrication, and Test , 2003 .

[17]  Andrew Gordon Wilson,et al.  GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration , 2018, NeurIPS.

[18]  Daniel M. Packwood,et al.  Bayesian Optimization for Materials Science , 2017 .

[19]  Andrew Gordon Wilson,et al.  When are Iterative Gaussian Processes Reliably Accurate? , 2021, ArXiv.

[20]  Peter I. Frazier,et al.  Bayesian Optimization of Composite Functions , 2019, ICML.