DynamoRep: Trajectory-Based Population Dynamics for Classification of Black-box Optimization Problems - Sorbonne Université Access content directly
Conference Papers Year : 2023

DynamoRep: Trajectory-Based Population Dynamics for Classification of Black-box Optimization Problems

Gjorgjina Cenikj
  • Function : Author
  • PersonId : 1148403
Gašper Petelin
  • Function : Author
  • PersonId : 1275691
Carola Doerr
Peter Korošec
  • Function : Author
  • PersonId : 1099544
Tome Eftimov
  • Function : Author
  • PersonId : 1099541

Abstract

The application of machine learning (ML) models to the analysis of optimization algorithms requires the representation of optimization problems using numerical features. These features can be used as input for ML models that are trained to select or to configure a suitable algorithm for the problem at hand. Since in pure blackbox optimization information about the problem instance can only be obtained through function evaluation, a common approach is to dedicate some function evaluations for feature extraction, e.g., using random sampling. This approach has two key downsides: (1) It reduces the budget left for the actual optimization phase, and (2) it neglects valuable information that could be obtained from a problem-solver interaction. In this paper, we propose a feature extraction method that describes the trajectories of optimization algorithms using simple descriptive statistics. We evaluate the generated features for the task of classifying problem classes from the Black Box Optimization Benchmarking (BBOB) suite. We demonstrate that the proposed DynamoRep features capture enough information to identify the problem class on which the optimization algorithm is running, achieving a mean classification accuracy of 95% across all experiments.
Fichier principal
Vignette du fichier
DynamoRep_GECCO_HAL.pdf (572.8 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04180581 , version 1 (12-08-2023)

Identifiers

Cite

Gjorgjina Cenikj, Gašper Petelin, Carola Doerr, Peter Korošec, Tome Eftimov. DynamoRep: Trajectory-Based Population Dynamics for Classification of Black-box Optimization Problems. GECCO '23: Genetic and Evolutionary Computation Conference, Jul 2023, Lisbon, Portugal. pp.813-821, ⟨10.1145/3583131.3590401⟩. ⟨hal-04180581⟩
22 View
9 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More