Searching of Self-similar Spaces

Kaliciak L., Myrhaug H., Goker A.

Future Technologies Conference (FTC), Vancouver, Kanada, 13 - 14 Kasım 2018, cilt.881, ss.1106-1124, (Tam Metin Bildiri) identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 881
  • Doi Numarası: 10.1007/978-3-030-02683-7_81
  • Basıldığı Şehir: Vancouver
  • Basıldığı Ülke: Kanada
  • Sayfa Sayıları: ss.1106-1124
  • Çanakkale Onsekiz Mart Üniversitesi Adresli: Hayır

Özet

In this paper we present a novel fractal encoding scheme for genetic algorithms based on iterated function systems. The algorithm is capable of encoding self-similar search spaces of fractional dimensions - including spaces of measure zero. Such self-similar spaces can naturally arise in many optimisation problems. In the paper, we also discuss the relationships between the Cantor set and probabilistic spaces, and the potential application of Cantor Dust as a combination of probability trees to create hybrid models. We conduct an experiment and report the results in order to illustrate the idea of fractal encoding. Finally, we also discuss the potential application areas of this new proposed algorithm.