Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/38775
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | MORALES HERNANDEZ, Alejandro | - |
dc.contributor.author | VAN NIEUWENHUYSE, Inneke | - |
dc.contributor.author | NAPOLES RUIZ, Gonzalo | - |
dc.date.accessioned | 2022-10-19T14:41:41Z | - |
dc.date.available | 2022-10-19T14:41:41Z | - |
dc.date.issued | 2022 | - |
dc.date.submitted | 2022-10-17T15:19:22Z | - |
dc.identifier.citation | Proceedings of International Conference on Optimization and Learning, | - |
dc.identifier.isbn | 978-3-031-22038-8 | - |
dc.identifier.isbn | 978-3-031-22039-5 | - |
dc.identifier.issn | 1865-0929 | - |
dc.identifier.uri | http://hdl.handle.net/1942/38775 | - |
dc.description.abstract | The performance of any Machine Learning (ML) algorithm is impacted by the choice of its hyperparameters. As training and evaluating a ML algorithm is usually expensive, the hyperparameter optimization (HPO) method needs to be computationally efficient to be useful in practice. Most of the existing approaches on multi-objective HPO use evolutionary strategies and metamodel-based optimization. However, few methods have been developed to account for uncertainty in the performance measurements. This paper presents results on multi-objective hyperparameter optimization with uncertainty on the evaluation of ML algorithms. We combine the sampling strategy of Tree-structured Parzen Estimators (TPE) with the metamodel obtained after training a Gaussian Process Regression (GPR) with heterogeneous noise. Experimental results on three analytical test functions and three ML problems show the improvement over multi-objective TPE and GPR, achieved with respect to the hypervolume indicator. | - |
dc.language.iso | en | - |
dc.publisher | SPRINGER INTERNATIONAL PUBLISHING AG | - |
dc.relation.ispartofseries | Communications in Computer and Information Science | - |
dc.subject.other | Hyperparameter optimization | - |
dc.subject.other | Multi-objective optimization | - |
dc.subject.other | Bayesian optimization | - |
dc.subject.other | Uncertainty | - |
dc.title | Multi-objective hyperparameter optimization with performance uncertainty | - |
dc.type | Proceedings Paper | - |
local.bibliographicCitation.conferencedate | 18/07/2022-20/07/2022 | - |
local.bibliographicCitation.conferencename | International Conference on Optimization and Learning | - |
local.bibliographicCitation.conferenceplace | Italy | - |
dc.identifier.epage | 46 | - |
dc.identifier.spage | 37 | - |
dc.identifier.volume | 1684 | - |
local.format.pages | 10 | - |
local.bibliographicCitation.jcat | C1 | - |
local.publisher.place | GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND | - |
local.type.refereed | Refereed | - |
local.type.specified | Proceedings Paper | - |
dc.identifier.doi | 10.1007/978-3-031-22039-5_4 | - |
dc.identifier.isi | 000917805300004 | - |
local.provider.type | - | |
local.bibliographicCitation.btitle | Proceedings of International Conference on Optimization and Learning | - |
local.uhasselt.international | yes | - |
item.contributor | MORALES HERNANDEZ, Alejandro | - |
item.contributor | VAN NIEUWENHUYSE, Inneke | - |
item.contributor | NAPOLES RUIZ, Gonzalo | - |
item.accessRights | Open Access | - |
item.fullcitation | MORALES HERNANDEZ, Alejandro; VAN NIEUWENHUYSE, Inneke & NAPOLES RUIZ, Gonzalo (2022) Multi-objective hyperparameter optimization with performance uncertainty. In: Proceedings of International Conference on Optimization and Learning,. | - |
item.fulltext | With Fulltext | - |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Peer review OLA2022.pdf Restricted Access | Conference material | 243.17 kB | Adobe PDF | View/Open Request a copy |
2209.04340.pdf | Non Peer-reviewed author version | 691.81 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.