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

Shane Carr, Roman Garnett, Cynthia Lo
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:898-907, 2016.

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.

Cite this Paper


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.} }
Endnote
%0 Conference Paper %T BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces %A Shane Carr %A Roman Garnett %A Cynthia Lo %B Proceedings of The 33rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Maria Florina Balcan %E Kilian Q. Weinberger %F pmlr-v48-carr16 %I PMLR %P 898--907 %U https://proceedings.mlr.press/v48/carr16.html %V 48 %X 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.
RIS
TY - CPAPER TI - BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces AU - Shane Carr AU - Roman Garnett AU - Cynthia Lo BT - Proceedings of The 33rd International Conference on Machine Learning DA - 2016/06/11 ED - Maria Florina Balcan ED - Kilian Q. Weinberger ID - pmlr-v48-carr16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 48 SP - 898 EP - 907 L1 - http://proceedings.mlr.press/v48/carr16.pdf UR - https://proceedings.mlr.press/v48/carr16.html AB - 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. ER -
APA
Carr, S., Garnett, R. & Lo, C.. (2016). BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces. Proceedings of The 33rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 48:898-907 Available from https://proceedings.mlr.press/v48/carr16.html.

Related Material