Publication:

{PF}²ES: Parallel Feasible Pareto Frontier Entropy Search for Multi-Objective Bayesian Optimization

Date

2023
Proceedings Paper
Simple item page Full metadata Statistics
 
dc.contributor.authorQing, Jixiang
dc.contributor.authorMoss, Henry B.
dc.contributor.authorDhaene, Tom
dc.contributor.authorCouckuyt, Ivo
dc.contributor.imecauthorQing, Jixiang
dc.contributor.imecauthorDhaene, Tom
dc.contributor.imecauthorCouckuyt, Ivo
dc.contributor.orcidimecQing, Jixiang::0000-0002-6446-6286
dc.contributor.orcidimecDhaene, Tom::0000-0003-2899-4636
dc.contributor.orcidimecCouckuyt, Ivo::0000-0002-9524-4205
dc.date.accessioned2024-12-03T10:20:23Z
dc.date.available2024-11-28T16:44:41Z
dc.date.available2024-11-29T10:56:43Z
dc.date.available2024-12-03T10:20:23Z
dc.date.embargo2023-04-30
dc.date.issued2023
dc.description.wosFundingTextThis research receives funding from the Flemish Government under the "Onderzoeksprogramma Artifciele Intelligentie (AI) Vlaanderen" programme and the "Fonds Wetenschappelijk Onderzoek (FWO)" programme, and Chinese Scholarship Council (grant number 201906290032). We sincerely thank Victor Picheny for the insightful comment on the necessity of considering the effect of spacing between discrete Pareto points, which has inspired proposition 1 and Algorithm 1. We also gratefully thank reviewers for providing extensive comments for improving the paper.
dc.identifier.issn2640-3498
dc.identifier.urihttps://imec-publications.be/handle/20.500.12860/44855
dc.publisherJMLR-JOURNAL MACHINE LEARNING RESEARCH
dc.source.beginpage2565
dc.source.conference26th International Conference on Artificial Intelligence and Statistics (AISTATS)
dc.source.conferencedateAPR 25-27, 2023
dc.source.conferencelocationValencia
dc.source.endpage2588
dc.source.journalProceedings of Machine Learning Research
dc.source.numberofpages24
dc.source.volume206
dc.subject.keywordsIMPROVEMENT
dc.subject.keywordsALGORITHM
dc.title

{PF}²ES: Parallel Feasible Pareto Frontier Entropy Search for Multi-Objective Bayesian Optimization

dc.typeProceedings paper
dspace.entity.typePublication
Files

Original bundle

Name:
8282.pdf
Size:
18.56 MB
Format:
Unknown data format
Description:
Published version
Download
Publication available in collections:
Conference contributions