Skip to main navigation Skip to search Skip to main content

BHO-MA: Bayesian Hyperparameter Optimization with Multi-objective Acquisition

Research output: Chapter in Book/Report/Conference proceedingsConference proceedingpeer-review

Abstract

Good hyperparameter values are crucial for the performance of machine learning models. In particular, poorly chosen values can cause under- or overfitting in regression and classification. A common approach to hyperparameter tuning is grid search, but this is crude and computationally expensive, and the literature contains several more efficient automatic methods such as Bayesian optimization. In this work, we develop a Bayesian hyperparameter optimization technique with more robust performance, by combining several acquisition functions and applying a multi-objective approach. We evaluated our method using both classification and regression tasks. We selected four data sets from the literature and compared the performance with eight popular methods. The results show that the proposed method achieved better results than all others.

Original languageEnglish
Title of host publicationOptimization, Learning Algorithms and Applications - 3rd International Conference, OL2A 2023, Revised Selected Papers
EditorsAna I. Pereira, Florbela P. Fernandes, Joao P. Coelho, Armando Mendes, Maria F. Pacheco, Jose Lima
PublisherSpringer Science and Business Media Deutschland GmbH
Pages391-408
Number of pages18
ISBN (Print)9783031530241
DOIs
Publication statusPublished - 2024
Event3rd International Conference on Optimization, Learning Algorithms and Applications, OL2A 2023 - Ponta Delgada, Portugal
Duration: 27 Sep 202329 Sep 2023

Publication series

NameCommunications in Computer and Information Science
Volume1981 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference3rd International Conference on Optimization, Learning Algorithms and Applications, OL2A 2023
Country/TerritoryPortugal
CityPonta Delgada
Period27/09/2329/09/23

UCC Futures

  • Artificial Intelligence and Data Analytics

Keywords

  • Bayesian Optimization
  • Hyperparameter Tuning
  • Multi-objective Optimization

Fingerprint

Dive into the research topics of 'BHO-MA: Bayesian Hyperparameter Optimization with Multi-objective Acquisition'. Together they form a unique fingerprint.

Cite this