Merge of evolutionary computation with gradient based method for optimization problems


Hewlett J., Wilamowski B., DÜNDAR G.

2007 IEEE International Symposium on Industrial Electronics, ISIE 2007, Caixanova - Vigo, İspanya, 4 - 07 Haziran 2007, ss.3304-3309, (Tam Metin Bildiri) identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/isie.2007.4375145
  • Basıldığı Şehir: Caixanova - Vigo
  • Basıldığı Ülke: İspanya
  • Sayfa Sayıları: ss.3304-3309
  • Boğaziçi Üniversitesi Adresli: Evet

Özet

The paper describes an optimization method which combines advantages of both evolutionary computation and gradient based methods. The proposed method follows the general concept of evolutionary computation, but uses an approximated gradient for generating subsequent populations. The gradient is not explicitly computed, but is instead estimated using minimum solutions from neighboring populations. Experimental data shows that the proposed method is not only robust, but is also comparable to gradient methods with respect to speed of convergence. ©2007 IEEE.