Automated Configuration of Genetic Algorithms by Tuning for Anytime Performance - Sorbonne Université
Journal Articles IEEE Transactions on Evolutionary Computation Year : 2022

Automated Configuration of Genetic Algorithms by Tuning for Anytime Performance

Abstract

Finding the best configuration of algorithms' hyperparameters for a given optimization problem is an important task in evolutionary computation. We compare in this work the results of four different hyperparameter optimization approaches for a family of genetic algorithms on 25 diverse pseudo-Boolean optimization problems. More precisely, we compare previously obtained results from a grid search with those obtained from three automated configuration techniques: iterated racing, mixedinteger parallel efficient global optimization, and mixed-integer evolutionary strategies. Using two different cost metrics, expected running time and the area under the empirical cumulative distribution function curve, we find that in several cases the best configurations with respect to expected running time are obtained when using the area under the empirical cumulative distribution function curve as the cost metric during the configuration process. Our results suggest that even when interested in expected running time performance, it might be preferable to use anytime performance measures for the configuration task. We also observe that tuning for expected running time is much more sensitive with respect to the budget that is allocated to the target algorithms.
Fichier principal
Vignette du fichier
TEVC-2106.06304.pdf (7.11 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03614648 , version 1 (21-03-2022)

Identifiers

Cite

Furong Ye, Carola Doerr, Hao Wang, Thomas Back. Automated Configuration of Genetic Algorithms by Tuning for Anytime Performance. IEEE Transactions on Evolutionary Computation, 2022, 26 (6), pp.1526 - 1538. ⟨10.1109/TEVC.2022.3159087⟩. ⟨hal-03614648⟩
51 View
39 Download

Altmetric

Share

More