Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/38775
Title: | Multi-objective hyperparameter optimization with performance uncertainty | Authors: | MORALES HERNANDEZ, Alejandro VAN NIEUWENHUYSE, Inneke NAPOLES RUIZ, Gonzalo |
Issue Date: | 2022 | Publisher: | SPRINGER INTERNATIONAL PUBLISHING AG | Source: | Proceedings of International Conference on Optimization and Learning, | Series/Report: | Communications in Computer and Information Science | 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. | Keywords: | Hyperparameter optimization;Multi-objective optimization;Bayesian optimization;Uncertainty | Document URI: | http://hdl.handle.net/1942/38775 | ISBN: | 978-3-031-22038-8 978-3-031-22039-5 |
DOI: | 10.1007/978-3-031-22039-5_4 | ISI #: | 000917805300004 | Category: | C1 | Type: | Proceedings Paper |
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.