Non-Elitist Selection Can Improve the Performance of Irace - Sorbonne Université
Conference Papers Year : 2022

Non-Elitist Selection Can Improve the Performance of Irace

Abstract

Modern optimization strategies such as evolutionary algorithms, ant colony algorithms, Bayesian optimization techniques, etc. come with several parameters that steer their behavior during the optimization process. To obtain high-performing algorithm instances, automated algorithm configuration techniques have been developed. One of the most popular tools is irace, which evaluates configurations in sequential races, making use of iterated statistical tests to discard poorly performing configurations. At the end of the race, a set of elite configurations are selected from those survivor configurations that were not discarded, using greedy truncation selection. We study two alternative selection methods: one keeps the best survivor and selects the remaining configurations uniformly at random from the set of survivors, while the other applies entropy to maximize the diversity of the elites. These methods are tested for tuning ant colony optimization algorithms for travelling salesperson problems and the quadratic assignment problem and tuning an exact tree search solver for satisfiability problems. The experimental results show improvement on the tested benchmarks compared to the default selection of irace. In addition, the obtained results indicate that non-elitist can obtain diverse algorithm configurations, which encourages us to explore a wider range of solutions to understand the behavior of algorithms.
Fichier principal
Vignette du fichier
IraceNonElitistSelection-PPSN2022.pdf (3.07 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03740761 , version 1 (29-07-2022)

Identifiers

Cite

Furong Ye, Diederick Vermetten, Carola Doerr, Thomas Bäck. Non-Elitist Selection Can Improve the Performance of Irace. 17th Proceedings of Parallel Problem Solving from Nature - (PPSN) 2022, Sep 2022, Dortmund, Germany. pp.32-45, ⟨10.1007/978-3-031-14714-2_3⟩. ⟨hal-03740761⟩
94 View
24 Download

Altmetric

Share

More