论文信息 - Different Learning Algorithms for Neural Networks - A Comparative Study

Different Learning Algorithms for Neural Networks - A Comparative Study

Neural Networks (NN) are usually trained with gradient search algorithms. Alternative approaches like genetic algorithms (GA) have been proposed before with promising results. In this paper six different training algorithms for NN are compared — two of them based on GA. The algorithms were evaluated with data from practically relevant applications of the Siemens AG.

Jochen Heistermann | J. Heistermann

[1] Otto Schmidbauer,et al. Classification of coarse phonetic categories in continuous speech: statistical classifiers vs. temporal flow connectionist network , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[2] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[3] H. P. Schwefel,et al. Numerische Optimierung von Computermodellen mittels der Evo-lutionsstrategie , 1977 .

[4] L. Darrell Whitley,et al. Optimizing Neural Networks Using FasterMore Accurate Genetic Search , 1989, ICGA.

[5] James L. McClelland,et al. Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[6] Jochen Heistermann. A parallel hybrid learning approach to artificial neural nets , 1991, Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing.