Beyond Landscape Analysis: DynamoRep Features For Capturing Algorithm-Problem Interaction In Single-Objective Continuous Optimization

The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem instances, for analyzing the behavior of optimization algorithms, and the quality of existing problem benchmarks, as well as for automated per-insta...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Evolutionary computation s. 1
Hlavní autoři: Cenikj, Gjorgjina, Petelin, Gašper, Doerr, Carola, Korošec, Peter, Eftimov, Tome
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States 07.03.2025
Témata:
algorithm classification
problem classification
single objective numerical optimization
feature construction
ISSN:1530-9304, 1530-9304
On-line přístup:Zjistit podrobnosti o přístupu
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem instances, for analyzing the behavior of optimization algorithms, and the quality of existing problem benchmarks, as well as for automated per-instance algorithm selection and configuration approaches. Extending purely problem-centered feature collections, our recently proposed DynamoRep features provide a simple and inexpensive representation of the algorithmproblem interaction during the optimization process. In this paper, we conduct a comprehensive analysis of the predictive power of the DynamoRep features for the problem classification, algorithm selection, and algorithm classification tasks. In particular, the features are evaluated for the classification of problem instances into problem classes from the BBOB (Black Box Optimization Benchmarking) suite, selecting the best algorithm to solve a given problem from a portfolio of three algorithms (Differential Evolution, Evolutionary Strategy, and Particle Swarm Optimization), as well as distinguishing these algorithms based on their trajectories. We show that, despite being much cheaper to compute, they can yield results comparable to those using state-ofthe-art Exploratory Landscape Analysis features.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1530-9304
1530-9304
DOI:10.1162/evco_a_00370