This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Sixth International Conference on Computational Intelligence and Multimedia Applications (ICCIMA'05)
Genetically Evolving Higher Order Neural Networks by Direct Encoding Method
Las Vegas, Nevada
August 16-August 18
ISBN: 0-7695-2358-7
Abdul Ahad Siddiqi, Karachi Institute of Information Technology
There are two major ways of encoding a neural network into a chromosome, as required in design of a Genetic Algorithm (GA). These are Explicit (direct) and Implicit (Indirect) encoding methods. The proposed direct encoding method to design Higher Order Neural Networks (HONN) does not use any known learning algorithm - rather it uses a gradient descent method to minimize the mean output error. The simple feed-forward network only uses one pass, called forward pass contrary to the standard learning algorithm which does the training in two passes. This saves an enormous amount of training time and the network converges to an optimum value as compared to other learning strategies.
Index Terms:
Genetic Algorithms, Neural Networks, Direct Encoding, Graph Grammars
Citation:
Abdul Ahad Siddiqi, "Genetically Evolving Higher Order Neural Networks by Direct Encoding Method," iccima, pp.62-67, Sixth International Conference on Computational Intelligence and Multimedia Applications (ICCIMA'05), 2005
Usage of this product signifies your acceptance of the Terms of Use.