dc.contributor.author | De Ath, G | |
dc.contributor.author | Everson, RM | |
dc.contributor.author | Rahat, AAM | |
dc.contributor.author | Fieldsend, JE | |
dc.date.accessioned | 2020-11-17T15:55:06Z | |
dc.date.issued | 2021-04-28 | |
dc.description.abstract | The performance of acquisition functions for Bayesian optimisation is investigated in terms of the Pareto front
between exploration and exploitation. We show that Expected Improvement and the Upper Confidence Bound
always select solutions to be expensively evaluated on the Pareto front, but Probability of Improvement is
never guaranteed to do so and Weighted Expected Improvement does only for a restricted range of weights.
We introduce two novel 𝜖-greedy acquisition functions. Extensive empirical evaluation of these together
with random search, purely exploratory and purely exploitative search on 10 benchmark problems in 1 to
10 dimensions shows that 𝜖-greedy algorithms are generally at least as effective as conventional acquisition
functions, particularly with a limited budget. In higher dimensions 𝜖-greedy approaches are shown to have
improved performance over conventional approaches. These results are borne out on a real world computational
fluid dynamics optimisation problem and a robotics active learning problem. | en_GB |
dc.description.sponsorship | Innovate UK | en_GB |
dc.identifier.citation | Vol. 1 (1), article 1 | en_GB |
dc.identifier.doi | 10.1145/3425501 | |
dc.identifier.grantnumber | 104400 | en_GB |
dc.identifier.uri | http://hdl.handle.net/10871/123653 | |
dc.language.iso | en | en_GB |
dc.publisher | Association for Computing Machinery (ACM) | en_GB |
dc.rights | © 2021 Copyright held by the owner/author(s). Publication rights licensed to ACM. | |
dc.subject | Bayesian optimisation | en_GB |
dc.subject | Acquisition function | en_GB |
dc.subject | Infill criteria | en_GB |
dc.subject | ε-greedy | en_GB |
dc.subject | Exploration-exploitation trade-off | en_GB |
dc.title | Greed is Good: Exploration and Exploitation Trade-offs in Bayesian Optimisation | en_GB |
dc.type | Article | en_GB |
dc.date.available | 2020-11-17T15:55:06Z | |
dc.description | This is the author accepted manuscript. The final version is available from ACM via the DOI in this record | en_GB |
dc.identifier.journal | ACM Transactions on Evolutionary Learning and Optimization (TELO) | en_GB |
dc.rights.uri | http://www.rioxx.net/licenses/all-rights-reserved | en_GB |
dcterms.dateAccepted | 2020-09-17 | |
exeter.funder | ::Innovate UK | en_GB |
rioxxterms.version | AM | en_GB |
rioxxterms.licenseref.startdate | 2020-09-17 | |
rioxxterms.type | Journal Article/Review | en_GB |
refterms.dateFCD | 2020-11-17T15:53:13Z | |
refterms.versionFCD | AM | |
refterms.dateFOA | 2021-05-05T15:21:18Z | |
refterms.panel | B | en_GB |