Title :
Generalization tools for the multilayer perceptron
Author :
Magnus, Amy L. ; Oxley, Mark E.
Author_Institution :
Intelligent Inf. Syst., Air Force Res. Lab., Wright-Patterson AFB, OH, USA
fDate :
6/24/1905 12:00:00 AM
Abstract :
A multilayer perceptron (MLP) represents a partial ordering over a feature set - an ordering based on the hyperplane arrangement implemented in the MLP´s first hidden layer. This paper presents tools for manipulating the partial ordering for better data generalization
Keywords :
backpropagation; data handling; generalisation (artificial intelligence); multilayer perceptrons; backpropagation; data generalization; feature set; halfspaces; hidden layer; hyperplane; multilayer perceptron; partial ordering; Helium; Information systems; Intelligent systems; Laboratories; Logic; Mathematics; Multilayer perceptrons; Nonhomogeneous media; Statistics; Training data;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007611
