%0 Conference Paper %1 Zhang-Muehlenbein-94-WCCI-EC %A Zhang, Byoung-Tak %A Mühlenbein, Heinz %B Proceedings of IEEE International Conference on Evolutionary Computation (ICEC-94), World Congress on Computational Intelligence %C Orlando, Florida, USA %D 1994 %I IEEE Computer Society Press %K algorithms, genetic programming %P 318--323 %T Synthesis of Sigma-Pi Neural Networks by the Breeder Genetic Programming %U http://bi.snu.ac.kr/Publications/Conferences/International/ICEC94.pdf %X Genetic programming has been successfully applied to evolve computer programs for solving a variety of interesting problems. In the previous work we introduced the breeder genetic programming (BGP) method that has Occam's razor in its fitness measure to evolve minimal size multilayer perceptrons. In this paper we apply the method to synthesis of sigma-pi neural networks. Unlike perceptron architectures, sigma-pi networks use product units as well as summation units to build higher-order terms. The effectiveness of the method is demonstrated on benchmark problems. Simulation results on noisy data suggest that BGP not only improves the generalization performance, it can also accelerate the convergence speed.

@inproceedings{Zhang-Muehlenbein-94-WCCI-EC, abstract = {Genetic programming has been successfully applied to evolve computer programs for solving a variety of interesting problems. In the previous work we introduced the breeder genetic programming (BGP) method that has Occam's razor in its fitness measure to evolve minimal size multilayer perceptrons. In this paper we apply the method to synthesis of sigma-pi neural networks. Unlike perceptron architectures, sigma-pi networks use product units as well as summation units to build higher-order terms. The effectiveness of the method is demonstrated on benchmark problems. Simulation results on noisy data suggest that BGP not only improves the generalization performance, it can also accelerate the convergence speed.}, added-at = {2008-06-19T17:46:40.000+0200}, address = {Orlando, Florida, USA}, author = {Zhang, Byoung-Tak and M{\"u}hlenbein, Heinz}, biburl = {https://www.bibsonomy.org/bibtex/20f459be7635932ca281da8e6641b8f22/brazovayeye}, booktitle = {Proceedings of IEEE International Conference on Evolutionary Computation (ICEC-94), World Congress on Computational Intelligence}, interhash = {3b88a0bc9c30e319904dbec673d51006}, intrahash = {0f459be7635932ca281da8e6641b8f22}, keywords = {algorithms, genetic programming}, month = {27-29 June}, notes = {Tests GP/Sigma-pi/NN on parity problems. On clean data was able to produce S/P Neural Networks with high performance >98% correct. Also ~90% on noisy data. Fitness function sums NN error and NN size/complexity penalty terms. Shows size/complexity penalty beneficial in that better NN are produced and the GP is twice as fast.}, pages = {318--323}, publisher = {IEEE Computer Society Press}, publisher_address = {New York, USA}, timestamp = {2008-06-19T17:55:13.000+0200}, title = {Synthesis of Sigma-Pi Neural Networks by the Breeder Genetic Programming}, url = {http://bi.snu.ac.kr/Publications/Conferences/International/ICEC94.pdf}, year = 1994 }