Title :
Evolving artificial neural network structures: experimental results for biologically-inspired adaptive mutations
Author :
Miller, Damon A. ; Arguello, Rodrigo ; Greenwood, Garrison W.
Author_Institution :
Dept. of Electr. & Comput. Eng., Western Michigan Univ., Kalamazoo, MI, USA
Abstract :
Previous work has suggested incorporating a biologically inspired mutation operator that reflects vertebrate neuron growth and death rates into an evolutionary algorithm for evolving artificial neural network structures. This paper further investigates this proposed approach by presenting experimental results for two classifier problems.
Keywords :
adaptive systems; evolutionary computation; neural nets; pattern classification; biologically-inspired adaptive mutations; classifier problems; evolutionary algorithm; evolving artificial neural network structures; mutation operator; neuron death rates; vertebrate neuron growth; Artificial neural networks; Biology computing; Computer networks; Evolutionary computation; Feedforward neural networks; Feedforward systems; Genetic mutations; Multi-layer neural network; Neural networks; Neurons;
Conference_Titel :
Evolutionary Computation, 2004. CEC2004. Congress on
Print_ISBN :
0-7803-8515-2
DOI :
10.1109/CEC.2004.1331157
