File(s) under permanent embargo
Exploiting strategy-space diversity for batch Bayesian optimisation
conference contribution
posted on 2018-01-01, 00:00 authored by Sunil GuptaSunil Gupta, Alistair ShiltonAlistair Shilton, Santu RanaSantu Rana, Svetha VenkateshSvetha VenkateshThis paper proposes a novel approach to batch Bayesian optimisation using a multi-objective optimisation framework with exploitation and exploration forming two objectives. The key advantage of this approach is that it uses a suite of strategies to balance exploration and exploitation and thus can efficiently handle the optimisation of a variety of functions with small to large number of local extrema. Another advantage is that it automatically determines the batch size within a specified budget avoiding unnecessary function evaluations. Theoretical analysis shows that the regret not only reduces sub-linearly but also by an additional reduction factor determined by the batch size. We demonstrate the efficiency of our algorithm by optimising a variety of benchmark functions, performing hyperparameter tuning of support vector regression and classification, and finally heat treatment process of an Al-Sc alloy. Comparisons with recent baseline algorithms confirm the usefulness of our algorithm.
History
Event
Machine Learning Research. Conference (21st : 2018 : Playa Blanca, Lanzarote, Canary Islands)Volume
84Series
Machine Learning Research ConferencePagination
538 - 547Publisher
Proceedings of Machine Learning ResearchLocation
Playa Blanca, Lanzarote, Canary IslandsPlace of publication
Cambridge, Mass.Start date
2018-04-09End date
2018-04-11Language
engPublication classification
E1 Full written paper - refereedCopyright notice
2018, the author(s)Editor/Contributor(s)
Amos Storkey, Fernando Perez-CruzTitle of proceedings
AISTATS 2018 : Proceedings of the International Conference on Artificial Intelligence and StatisticsUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC