Pregled bibliografske jedinice broj: 81370
Sensitivity of Differential Evolution Algorithm to Values of Control Parameters
Sensitivity of Differential Evolution Algorithm to Values of Control Parameters // Proceedings of the International Conference on Artificial Intelligence / Arabnia, H.R.; Mun, Y. (ur.).
Las Vegas (NV), Sjedinjene Američke Države: CSREA Press, 2002. str. 1087-1093 (poster, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 81370 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Sensitivity of Differential Evolution Algorithm to Values of Control Parameters
Autori
Šmuc, Tomislav
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the International Conference on Artificial Intelligence
/ Arabnia, H.R.; Mun, Y. - : CSREA Press, 2002, 1087-1093
Skup
International Conference on Artificial Intelligence
Mjesto i datum
Las Vegas (NV), Sjedinjene Američke Države, 24.06.2002. - 27.06.2002
Vrsta sudjelovanja
Poster
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
differential evolutionary algorithm; convergence; sensitivity analysis
Sažetak
Differential evolution is a relatively novel evolutionary algorithm for solving general type of non-linear, constrained optimization problems. Its main advantages are simplicity and good convergence properties, at least for smaller and medium sized problems. Although convergence properties of differential evolution algorithm for modestly dimensioned problems reported in the literature were judged as excellent, analyses of algorithm's performance are still rather elementary. This work presents results of a detailed sensitivity analysis of control parameters on the suite of optimization problems. Relative importance of control parameters, the sensitivity of different strategies to problem types and possible improvements and directions of research are drawn from the obtained results.
Izvorni jezik
Engleski
Znanstvena područja
Elektrotehnika
POVEZANOST RADA