Impact of Training Instance Selection on Automated Algorithm Selection Models for Numerical Black-box Optimization - Sorbonne Université Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Impact of Training Instance Selection on Automated Algorithm Selection Models for Numerical Black-box Optimization

Résumé

The recently proposed MA-BBOB function generator provides a way to create numerical black-box benchmark problems based on the well-established BBOB suite. Initial studies on this generator highlighted its ability to smoothly transition between the component functions, both from a low-level landscape feature perspective, as well as with regard to algorithm performance. This suggests that MA-BBOB-generated functions can be an ideal testbed for automated machine learning methods, such as automated algorithm selection (AAS). In this paper, we generate 11\,800 functions in dimensions $d=2$ and $d=5$, respectively, and analyze the potential gains from AAS by studying performance complementarity within a set of eight algorithms. We combine this performance data with exploratory landscape features to create an AAS pipeline that we use to investigate how to efficiently select training sets within this space. We show that simply using the BBOB component functions for training yields poor test performance, while the ranking between uniformly chosen and diversity-based training sets strongly depends on the distribution of the test set.

Mots clés

Fichier principal
Vignette du fichier
GECCO-2024-InstanceSelection-MABBOB.pdf (4.17 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04580570 , version 1 (20-05-2024)

Identifiants

Citer

Konstantin Dietrich, Diederick Vermetten, Carola Doerr, Pascal Kerschke. Impact of Training Instance Selection on Automated Algorithm Selection Models for Numerical Black-box Optimization. GECCO '24: Proceedings of the Genetic and Evolutionary Computation Conference, ACM, Jul 2024, Melbourne, Australia. ⟨10.1145/3638529.3654100⟩. ⟨hal-04580570⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More