Is greed still good in multi-objective Bayesian optimisation?
dc.contributor.author | Leite Richardson, F | |
dc.contributor.author | De Ath, G | |
dc.contributor.author | Chugh, T | |
dc.date.accessioned | 2024-07-02T08:56:55Z | |
dc.date.issued | 2024 | |
dc.date.updated | 2024-07-01T16:21:38Z | |
dc.description.abstract | Bayesian optimisation (BO) is a popular tool for solving expensive optimisation problems. BO utilises Bayesian models and balances exploitation and exploration in searching for potential solutions. In this work, we investigate the trade-off between exploration and exploitation in multi-objective BO by comparing three different approaches: selecting points on the estimated Pareto front (PF) of the predicted values of the surrogate models, selecting points on the estimated PF of the predicted uncertainty of the models, and using an \egreedy approach to balance between the two PFs. We evaluate the performance of these approaches on a set of benchmark problems and compare them to a random baseline and expected hypervolume improvement (EHVI). It was found that the \egreedy and fully exploitative approaches were the best performing across all problem dimensionalities and that the performance of EHVI comparatively decreased as the dimensionality of the problem increased. | en_GB |
dc.description.sponsorship | Engineering and Physical Sciences Research Council (EPSRC) | en_GB |
dc.identifier.citation | Genetic and Evolutionary Computation Conference (GECCO'24), Melbourne, Australia, 14 - 18 July 2024. Awaiting full citation and resolution of DOI | en_GB |
dc.identifier.doi | https://doi.org/10.1145/3638530.3664189 | |
dc.identifier.grantnumber | EP/V056522/1 | en_GB |
dc.identifier.uri | http://hdl.handle.net/10871/136535 | |
dc.identifier | ORCID: 0000-0003-4909-0257 (De Ath, George) | |
dc.identifier | ScopusID: 57205564689 (De Ath, George) | |
dc.identifier | ResearcherID: AAP-8110-2021 (De Ath, George) | |
dc.language.iso | en | en_GB |
dc.publisher | Association for Computing Machinery (ACM) | en_GB |
dc.rights.embargoreason | Under temporary indefinite embargo pending publication by ACM. No embargo required on publication | en_GB |
dc.rights | For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission | en_GB |
dc.subject | Surrogate modelling | en_GB |
dc.subject | Gaussian process | en_GB |
dc.subject | Multi-objective optimisation | en_GB |
dc.subject | Greed is good | en_GB |
dc.subject | Epsilon-greedy | en_GB |
dc.subject | Uncertainty quantification | en_GB |
dc.title | Is greed still good in multi-objective Bayesian optimisation? | en_GB |
dc.type | Conference paper | en_GB |
dc.date.available | 2024-07-02T08:56:55Z | |
dc.identifier.isbn | 979-8-4007-0495-6 | |
exeter.location | Melbourne, VIC, Australia | |
dc.description | This is the author accepted manuscript. | en_GB |
dc.relation.ispartof | Genetic and Evolutionary Computation Conference (GECCO '24 Companion) | |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | en_GB |
dcterms.dateAccepted | 2024-05-03 | |
dcterms.dateSubmitted | 2024-04-12 | |
rioxxterms.version | AM | en_GB |
rioxxterms.licenseref.startdate | 2024-05-03 | |
rioxxterms.type | Conference Paper/Proceeding/Abstract | en_GB |
refterms.dateFCD | 2024-07-01T16:21:40Z | |
refterms.versionFCD | AM | |
refterms.panel | B | en_GB |
pubs.name-of-conference | Genetic and Evolutionary Computation Conference | |
exeter.rights-retention-statement | Yes |
Files in this item
This item appears in the following Collection(s)
Except where otherwise noted, this item's licence is described as For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission