论文信息 - A Study and Improvement of the Genetic Algorithm in the CAM-Brain Machine

A Study and Improvement of the Genetic Algorithm in the CAM-Brain Machine

This paper presents a study of the CAM–Brain Machine (CBM), a hardware tool which implements a cellular automata based neural network. The idea of this machine is to build a brain, consisting of up to 64,640 modules. Each of these modules implements a neural network with up to 1152 neurons. The structure of these modules is not fixed, but evolves directly in hardware under the control of a built-in genetic algorithm that guides the evolution. The goal was to analyse this existing genetic algorithm in the CBM, discover some of its weaknesses and present a better alternative.

Yvan Saeys | H. Van Marck

[1] William Bialek,et al. Spikes: Exploring the Neural Code , 1996 .

[2] Christopher J. Bishop,et al. Pulsed Neural Networks , 1998 .

[3] Hugo de Garis,et al. "CBM (CAM-BRAIN MACHINE)" A Hardware Tool which Evolves a Neural Net Module in a Fraction of a Second and Runs a Million Neuron Artificial Brain in Real Time , 1997 .

[4] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .