caa a22 4500 475800982 CHVBK 20180406123737.0 cr unu---uuuuu 170329e20001001xx s 000 0 eng 10.1023/A:1009684907680 doi (NATIONALLICENCE)springer-10.1023/A:1009684907680 Evolving Multilayer Perceptrons [Elektronische Daten] [P. Castillo, J. Carpio, J. Merelo, A. Prieto, V. Rivas, G. Romero] This paper proposes a new version of a method (G-Prop, genetic backpropagation) that attempts to solve the problem of finding appropriate initial weights and learning parameters for a single hidden layer Multilayer Perceptron (MLP) by combining an evolutionary algorithm (EA) and backpropagation (BP). The EA selects the MLP initial weights, the learning rate and changes the number of neurons in the hidden layer through the application of specific genetic operators, one of which is BP training. The EA works on the initial weights and structure of the MLP, which is then trained using QuickProp; thus G-Prop combines the advantages of the global search performed by the EA over the MLP parameter space and the local search of the BP algorithm. The application of the G-Prop algorithm to several real-world and benchmark problems shows that MLPs evolved using G-Prop are smaller and achieve a higher level of generalization than other perceptron training algorithms, such as QuickPropagation or RPROP, and other evolutive algorithms. It also shows some improvement over previous versions of the algorithm. Kluwer Academic Publishers, 2000 evolutionary algorithms nationallicence generalization nationallicence learning nationallicence neural networks nationallicence optimization nationallicence Castillo P. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut Carpio J. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut Merelo J. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut Prieto A. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut Rivas V. Department of Computer Science, University of Jaén, Avda. Madrid, 35, E. 23071, Jaén, Spain aut Romero G. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut Neural Processing Letters Kluwer Academic Publishers 12/2(2000-10-01), 115-128 1370-4621 12:2<115 2000 12 11063 https://doi.org/10.1023/A:1009684907680 text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1023/A:1009684907680 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Castillo P. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut NATIONALLICENCE 700 1- Carpio J. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut NATIONALLICENCE 700 1- Merelo J. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut NATIONALLICENCE 700 1- Prieto A. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut NATIONALLICENCE 700 1- Rivas V. Department of Computer Science, University of Jaén, Avda. Madrid, 35, E. 23071, Jaén, Spain aut NATIONALLICENCE 700 1- Romero G. Department of Architecture and Computer Technology, University of Granada, Campus de Fuentenueva, E. 18071, Granada, Spain aut NATIONALLICENCE 773 0- Neural Processing Letters Kluwer Academic Publishers 12/2(2000-10-01), 115-128 1370-4621 12:2<115 2000 12 11063 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer