ICML 2016poster29 citations

BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces

Shane Carr, Roman Garnett, Cynthia Lo

Abstract

We present a novel application of Bayesian optimization to the field of surface science: rapidly and accurately searching for the global minimum on potential energy surfaces. Controlling molecule-surface interactions is key for applications ranging from environmental catalysis to gas sensing. We present pragmatic techniques, including exploration/exploitation scheduling and a custom covariance kernel that encodes the properties of our objective function. Our method, the Bayesian Active Site Calculator (BASC), outperforms differential evolution and constrained minima hopping – two state-of-the-art approaches – in trial examples of carbon monoxide adsorption on a hematite substrate, both with and without a defect.

BibTeX
@InProceedings{pmlr-v48-carr16,
  title = 	 {BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces},
  author = 	 {Carr, Shane and Garnett, Roman and Lo, Cynthia},
  booktitle = 	 {Proceedings of The 33rd International Conference on Machine Learning},
  pages = 	 {898--907},
  year = 	 {2016},
  editor = 	 {Balcan, Maria Florina and Weinberger, Kilian Q.},
  volume = 	 {48},
  series = 	 {Proceedings of Machine Learning Research},
  address = 	 {New York, New York, USA},
  month = 	 {20--22 Jun},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v48/carr16.pdf},
  url = 	 {https://proceedings.mlr.press/v48/carr16.html},
  abstract = 	 {We present a novel application of Bayesian optimization to the field of surface science: rapidly and accurately searching for the global minimum on potential energy surfaces. Controlling molecule-surface interactions is key for applications ranging from environmental catalysis to gas sensing. We present pragmatic techniques, including exploration/exploitation scheduling and a custom covariance kernel that encodes the properties of our objective function. Our method, the Bayesian Active Site Calculator (BASC), outperforms differential evolution and constrained minima hopping – two state-of-the-art approaches – in trial examples of carbon monoxide adsorption on a hematite substrate, both with and without a defect.}
}