Global Optimization Using Meta-Controlled Boltzmann Machine

Author

Yaakob, Shamshul Bahar ; Watada, Junzo

Author_Institution

Grad. Sch. of IPS, Waseda Univ., Kitakyushu, Japan

fYear

2010

fDate

13-15 Dec. 2010

Firstpage

39

Lastpage

42

Abstract

In this study, a new artificial neuron network model called the meta-controlled Boltzmann machine is introduced. The meta-controlled Boltzmann machine model includes the McCulloch-Pitts model, the Hop field network, and also the Boltzmann machine. The proposed method are applied both diffusion processes and simulated annealing. The convergence proof of the proposed method is shows in this paper. Meta-controlled Boltzmann machine show an ability to solve combinatorial optimization problems better than either Hop field networks or Boltzmann machines.

Keywords

Hopfield neural nets; combinatorial mathematics; diffusion; simulated annealing; Hop field network; McCulloch-Pitts model; artificial neuron network model; combinatorial optimization problems; diffusion processes; global optimization; meta-controlled Boltzmann machine model; simulated annealing; Artificial neural networks; Equations; Hopfield neural networks; Mathematical model; Noise; Optimization; Substations; Boltzmann machine; Hopfield networks; Neural network; Simulated annealing; meta-controlled Boltzmann machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Genetic and Evolutionary Computing (ICGEC), 2010 Fourth International Conference on

Conference_Location

Shenzhen

Print_ISBN

978-1-4244-8891-9

Electronic_ISBN

978-0-7695-4281-2

Type

conf

DOI

10.1109/ICGEC.2010.18

Filename

5715365