Selecting the selector: Comparison of update rules for discrete global optimization

We compare some well-known Bayesian global optimization methods in four distinct regimes, corresponding to high and low levels of measurement noise and to high and low levels of “quenched noise” (which term we use to describe the roughness of the function we are trying to optimize). We isolate the two stages of this optimization in terms of a “regressor,” which fits a model to the data measured so far, and a “selector,” which identifies the next point to be measured. The focus of this paper is to investigate the choice of selector when the regressor is well matched to the data.

[1]  N. Zheng,et al.  Global Optimization of Stochastic Black-Box Systems via Sequential Kriging Meta-Models , 2006, J. Glob. Optim..

[2]  Victor Picheny,et al.  Quantile-Based Optimization of Noisy Computer Experiments With Tunable Precision , 2013, Technometrics.

[3]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[4]  D. Ginsbourger,et al.  A benchmark of kriging-based infill criteria for noisy optimization , 2013, Structural and Multidisciplinary Optimization.

[5]  Robert B. Gramacy,et al.  Cases for the nugget in modeling computer experiments , 2010, Statistics and Computing.

[6]  S. Gupta,et al.  Bayesian look ahead one-stage sampling allocations for selection of the best population , 1996 .

[7]  Warren B. Powell,et al.  A Knowledge-Gradient Policy for Sequential Information Collection , 2008, SIAM J. Control. Optim..

[8]  Philipp Hennig,et al.  Entropy Search for Information-Efficient Global Optimization , 2011, J. Mach. Learn. Res..

[9]  James Theiler,et al.  Adaptive Strategies for Materials Design using Uncertainties , 2016, Scientific Reports.

[10]  James Theiler,et al.  Accelerated search for materials with targeted properties by adaptive design , 2016, Nature Communications.

[11]  Pritam Ranjan,et al.  Comment: EI Criteria for Noisy Computer Simulators , 2013, Technometrics.

[12]  Jason L. Loeppky,et al.  Batch sequential design to achieve predictive maturity with calibrated computer models , 2011, Reliab. Eng. Syst. Saf..

[13]  Olivier Dubrule,et al.  Cross validation of kriging in a unique neighborhood , 1983 .

[14]  James Theiler,et al.  A Perspective on Materials Informatics: State-of-the-Art and Challenges , 2016 .

[15]  J. Mockus,et al.  The Bayesian approach to global optimization , 1989 .

[16]  Chiho Kim,et al.  From Organized High-Throughput Data to Phenomenological Theory using Machine Learning: The Example of Dielectric Breakdown , 2016 .

[17]  Harold J. Kushner,et al.  A New Method of Locating the Maximum Point of an Arbitrary Multipeak Curve in the Presence of Noise , 1964 .

[18]  Rampi Ramprasad,et al.  Computational strategies for polymer dielectrics design , 2014 .

[19]  X. Emery.  The kriging update equations and their application to the selection of neighboring data , 2009 .

[20]  Arun Mannodi-Kanakkithodi,et al.  Machine Learning Strategy for Accelerated Design of Polymer Dielectrics , 2016, Scientific Reports.

[21]  Warren B. Powell,et al.  The Knowledge-Gradient Policy for Correlated Normal Beliefs , 2009, INFORMS J. Comput..

[22]  Eric Walter,et al.  An informational approach to the global optimization of expensive-to-evaluate functions , 2006, J. Glob. Optim..