TY - GEN A1 - Williams, Paul A1 - Duller, Andrew W.G. A2 - Korbicz, Józef - red. A2 - Uciński, Dariusz - red. PB - Zielona Góra: Uniwersytet Zielonogórski N2 - The back-propagation algorithm, for training multi-layer perceptrons tends to be slow to converge to a final solution and many methods have been proposed for improving this. One technique takes advantage of an alternative training error criterion, however, we show that this reduces the robustness of the learning in the presence of outliers in the input data. Two examples are used to show the characteristics of the learning methods, one a test problem and the other from a "real-world" problem. L1 - http://zbc.uz.zgora.pl/repozytorium/Content/57710/AMCS_1995_5_4_10.pdf L2 - http://zbc.uz.zgora.pl/repozytorium/Content/57710 KW - sterowanie KW - sterowanie-teoria KW - sztuczna inteligencja KW - matematyka stosowana KW - informatyka T1 - On the robustness of learning in the multi-layer perceptron UR - http://zbc.uz.zgora.pl/repozytorium/dlibra/docmetadata?id=57710 ER -