hrcak mascot   Srce   HID

Izvorni znanstveni članak

Convergence Theory and Applications of the Factorized Distribution Algorithm

Heinz Muhlenbein ; RWCP Theoretical Foundation GMO Laboratory, GMO, Forschungszentrum lnlormationstechnik, St. Augustin
Thilo Mahnig ; RWCP Theoretical Foundation GMO Laboratory, GMO, Forschungszentrum lnlormationstechnik, St. Augustin

Puni tekst: engleski, pdf (6 MB) str. 19-32 preuzimanja: 85* citiraj
APA 6th Edition
Muhlenbein, H. i Mahnig, T. (1999). Convergence Theory and Applications of the Factorized Distribution Algorithm. Journal of computing and information technology, 7 (1), 19-32. Preuzeto s https://hrcak.srce.hr/150197
MLA 8th Edition
Muhlenbein, Heinz i Thilo Mahnig. "Convergence Theory and Applications of the Factorized Distribution Algorithm." Journal of computing and information technology, vol. 7, br. 1, 1999, str. 19-32. https://hrcak.srce.hr/150197. Citirano 16.07.2019.
Chicago 17th Edition
Muhlenbein, Heinz i Thilo Mahnig. "Convergence Theory and Applications of the Factorized Distribution Algorithm." Journal of computing and information technology 7, br. 1 (1999): 19-32. https://hrcak.srce.hr/150197
Harvard
Muhlenbein, H., i Mahnig, T. (1999). 'Convergence Theory and Applications of the Factorized Distribution Algorithm', Journal of computing and information technology, 7(1), str. 19-32. Preuzeto s: https://hrcak.srce.hr/150197 (Datum pristupa: 16.07.2019.)
Vancouver
Muhlenbein H, Mahnig T. Convergence Theory and Applications of the Factorized Distribution Algorithm. Journal of computing and information technology [Internet]. 1999 [pristupljeno 16.07.2019.];7(1):19-32. Dostupno na: https://hrcak.srce.hr/150197
IEEE
H. Muhlenbein i T. Mahnig, "Convergence Theory and Applications of the Factorized Distribution Algorithm", Journal of computing and information technology, vol.7, br. 1, str. 19-32, 1999. [Online]. Dostupno na: https://hrcak.srce.hr/150197. [Citirano: 16.07.2019.]

Sažetak
The paper investigates the optimization of additively decomposable functions (ADF) by a new evolutionary algorithm called Factorized Distribution Algorithm (FDA). FDA is based on a factorization of the distribution to generate search points. First separable ADFs are considered. These are mapped to generalized linear functions with metavariables defined for multiple alleles. The mapping transforms FDA into an Univariate Marginal Frequency Algorithm (UMDA). For UMDA the exact equation for the response to selection is.computed under the assumption of proportionate selection. For truncation selection an approximate equation for the time to convergence is used, derived from an analysis of the OneMax function. FDA is also numerically investigated for non separable functions. The time to convergence is very similar to separable ADFs. FDA outpe1iorms the genetic algorithm with recombination of strings by far.

Ključne riječi
response to selection; Fisher's Theorem; additively decomposed functions; genetic algorithm; factorized di stribution

Hrčak ID: 150197

URI
https://hrcak.srce.hr/150197

Posjeta: 150 *