berk-randomisedgaussian-2021.pdf (7.99 MB)
Randomised Gaussian Process Upper Confidence Bound for Bayesian Optimisation
conference contribution
posted on 2020-01-01, 00:00 authored by Julian BerkJulian Berk, Sunil GuptaSunil Gupta, Santu RanaSantu Rana, Svetha VenkateshSvetha VenkateshIn order to improve the performance of Bayesian optimisation, we develop a modified Gaussian process upper confidence bound (GP-UCB) acquisition function. This is done by sampling the exploration-exploitation trade-off parameter from a distribution. We prove that this allows the expected trade-off parameter to be altered to better suit the problem without compromising a bound on the function's Bayesian regret. We also provide results showing that our method achieves better performance than GP-UCB in a range of real-world and synthetic problems.
History
Event
Artificial Intelligence and Seventeenth Pacific Rim International. Joint Conference on Artificial Intelligence (2021 : 29th : Yokohama, Japan)Pagination
2284 - 2290Publisher
The ConferenceLocation
Yokohama, JapanPlace of publication
[Yokohama, Japan]Publisher DOI
Link to full text
Start date
2021-01-07End date
2021-01-15ISBN-13
9780999241165Language
engPublication classification
E1 Full written paper - refereedTitle of proceedings
IJCAI-PRICAI-20 : Proceedings of the Twenty-Ninth International Joint Conference on Artificial IntelligenceUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC