RNDr. Marko Genyk-Berezovskyj

Všechny publikace

Hybrid Inductive Models: Deterministic Crowding Employed

  • Autoři: Kordík, P., Šnorek, M., RNDr. Marko Genyk-Berezovskyj,
  • Publikace: Proceedings of the International Joint Conference on Neural Networks. Piscataway: IEEE, 2004. p. 2343-2346. ISBN 0-7803-8360-5.
  • Rok: 2004
  • Pracoviště: Katedra počítačů
  • Anotace:
    Our research draws on experience with Group Method of Data Handling (GMDH) introduced by Ivachknenko in 1966. We have modified Multilayered Iterative Algorithm (MIA) that is commonly used to generate inductive models of real-world systems. In our algorithm heterogeneous units are used instead of units with given polynomial transfer function and therefore Hybrid Inductive Models (HIMs) are generated. This paper shows how to improve the efficiency of search for optimal HIMs. This is attained by employing Deterministic Crowding (DC) method proposed by Mahfoud in 1995. As a by-product of using DC method, we can estimate the importance of input variables for modeled output (sensitivity analysis).

A Modified GMDH Method and Model Quality Evaluation by Visualization

Global Team Teaching

The Modified GMDH Method Applied to Model Complex Systems

  • Autoři: Kordík, P., Ing. Pavel Náplava, Ph.D., Šnorek, M., RNDr. Marko Genyk-Berezovskyj,
  • Publikace: International Conference on Inductive Modeling - ICIM 2002. Lviv: State Scientific and Research Institute of Information Infrastructure, 2002, pp. 150-155. ISSN 0135-5465.
  • Rok: 2002

Tools for Distributed Learning at the University

Za stránku zodpovídá: Ing. Mgr. Radovan Suk