dc.contributor.author | De Ath, G | |
dc.contributor.author | Everson, RM | |
dc.contributor.author | Fieldsend, JE | |
dc.date.accessioned | 2021-08-06T15:57:13Z | |
dc.date.issued | 2021-12-01 | |
dc.description.abstract | Batch Bayesian optimisation (BO) is a successful technique for the optimisation of expensive black-box functions. Asynchronous BO can reduce wallclock time by starting a new evaluation as soon as another finishes, thus maximising resource utilisation. To maximise resource allocation, we develop a novel asynchronous BO method, AEGiS (Asynchronous ε-Greedy Global Search) that combines greedy search, exploiting the surrogate's mean prediction, with Thompson sampling and random selection from the approximate Pareto set describing the trade-off between exploitation (surrogate mean prediction) and exploration (surrogate posterior variance). We demonstrate empirically the efficacy of AEGiS on synthetic benchmark problems, meta-surrogate hyperparameter tuning problems and real-world problems, showing that AEGiS generally outperforms existing methods for asynchronous BO. When a single worker is available performance is no worse than BO using expected improvement. | en_GB |
dc.description.sponsorship | Innovate UK | en_GB |
dc.identifier.citation | Vol. 161, pp. 578–588 | en_GB |
dc.identifier.grantnumber | 104400 | en_GB |
dc.identifier.uri | http://hdl.handle.net/10871/126698 | |
dc.language.iso | en | en_GB |
dc.publisher | ML Research Press | en_GB |
dc.relation.url | https://proceedings.mlr.press/v161/de-ath21a.html | |
dc.rights | © 2021 Author(s) | |
dc.title | Asynchronous ε-greedy Bayesian optimisation | en_GB |
dc.type | Conference paper | en_GB |
dc.date.available | 2021-08-06T15:57:13Z | |
dc.identifier.issn | 2640-3498 | |
dc.identifier.issn | 2640-3498 | |
dc.description | This is the author accepted manuscript. The final version is available from ML Research Press via the link in this record | en_GB |
dc.description | UAI2021: 37th Conference on Uncertainty in Artificial Intelligence, 27 - 30 July 2021. Online | |
dc.identifier.journal | Proceedings of Machine Learning Research (PMLR) | |
dc.rights.uri | http://www.rioxx.net/licenses/all-rights-reserved | en_GB |
dcterms.dateAccepted | 2021-05-12 | |
exeter.funder | ::Innovate UK | en_GB |
exeter.funder | ::Innovate UK | en_GB |
rioxxterms.version | AM | en_GB |
rioxxterms.licenseref.startdate | 2021-05-12 | |
rioxxterms.type | Conference Paper/Proceeding/Abstract | en_GB |
refterms.dateFCD | 2021-08-06T15:38:28Z | |
refterms.versionFCD | AM | |
refterms.dateFOA | 2022-01-06T14:55:19Z | |
refterms.panel | B | en_GB |