Bayesian optimisation is an efficient method for global optimisation of expensive black-box functions. However, the current Gaussian process based methods cater to functions with arbitrary smoothness, and do not explicitly model the fact that most of the real world optimisation problems are well-behaved functions with only a few peaks. In this paper, we incorporate such shape constraints through the use of a derivative meta-model. The derivative meta-model is built using a Gaussian process with a polynomial kernel and derivative samples from this meta-model are used as extra observations to the standard Bayesian optimisation procedure. We provide a Bayesian framework to infer the degree of the polynomial kernel. Experiments on both benchmark functions and hyperparameter tuning problems demonstrate the superiority of our approach over baselines.
History
Volume
11013
Pagination
256-264
Location
Nanjing, China
Start date
2018-08-28
End date
2018-08-31
ISSN
0302-9743
eISSN
1611-3349
ISBN-13
9783319973098
Language
eng
Publication classification
E1 Full written paper - refereed
Copyright notice
2018, Springer International Publishing AG, part of Springer Nature
Editor/Contributor(s)
Geng X, Kang BH
Title of proceedings
PRICAI 2018: Proceedings of the 15th Pacific Rim International Conference on Artificial Intelligence
Event
Jiangsu Association of Artificial Intelligence. Conference (15th : 2018 : Nanjing, China)
Publisher
Springer
Place of publication
Cham, Switzerland
Series
Jiangsu Association of Artificial Intelligence Conference