On measuring and improving the quality of linkage learning in modern evolutionary algorithms applied to solve partially additively separable problems
| Autor: | Marcin M. Komarnicki, Bartosz Frej, Michal W. Przewozniczek |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Dependency (UML)
Computer science business.industry media_common.quotation_subject Evolutionary algorithm 0102 computer and information sciences 02 engineering and technology Linkage (mechanical) Machine learning computer.software_genre Design structure matrix 01 natural sciences law.invention Estimation of distribution algorithm 010201 computation theory & mathematics law Genetic algorithm 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Quality (business) Artificial intelligence business computer media_common |
| Zdroj: | GECCO |
| DOI: | 10.1145/3377930.3390242 |
| Popis: | Linkage learning is frequently employed in modern evolutionary algorithms. High linkage quality may be the key to an evolutionary method's effectiveness. Similarly, the faulty linkage may be the reason for its poor performance. Many state-of-the-art evolutionary methods use a Dependency Structure Matrix (DSM) to obtain linkage. In this paper, we propose a quality measure for DSM. Based on this measure, we analyze the behavior of modern evolutionary methods. We show the dependency between the linkage and the effectiveness of the considered methods. Finally, we propose a framework that improves the quality of the linkage. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|