论文信息 - Minimizing the System Error in Feedforward Neural Networks with Evolution Strategy

Minimizing the System Error in Feedforward Neural Networks with Evolution Strategy

In this paper evolution strategy is applied to minimize the system error in feedforward neural networks. The evolution strategy does not use externally tuned learning parameters. Moreover, it is not necessary to evaluate a gradient information as required by the backpropagation algorithm. Experimental results are presented and compared with the standard backpropagation technique.

Willfried Wienholt

[1] Hans-Paul Schwefel,et al. Numerical optimization of computer models , 1981 .

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

[3] Andreas Ostermeier,et al. An Evolution Strategy with Momentum Adaptation of the Random Number Distribution , 1992, PPSN.

[4] Ingo Rechenberg,et al. The Evolution Strategy. A Mathematical Model of Darwinian Evolution , 1984 .

[5] Robert A. Jacobs,et al. Increased rates of convergence through learning rate adaptation , 1987, Neural Networks.

[6] Roger J.-B. Wets,et al. Minimization by Random Search Techniques , 1981, Math. Oper. Res..

[7] Tom Tollenaere,et al. SuperSAB: Fast adaptive back propagation with good scaling properties , 1990, Neural Networks.

[8] Norio Baba,et al. A new approach for finding the global minimum of error function of neural networks , 1989, Neural Networks.