Title :
Delta learning law for a single neuron
Author_Institution :
Dept. of Civil Eng., Indian Inst. of Technol., Madras, India
Abstract :
In this paper we have suggested a modification to the delta learning law. The modification is essentially the addition of terms having higher order derivatives to the term containing the first order derivative in the conventional delta leaning law. Faster convergence is achieved in the experiments conducted on the XOR problem
Keywords :
convergence; feedforward neural nets; learning (artificial intelligence); multilayer perceptrons; MLFFNN; XOR problem; convergence; delta learning law; first-order derivative; high-order derivatives; multilayer feedforward neural network; neuron; Civil engineering; Convergence; Feedforward neural networks; Joining processes; Logic functions; Multi-layer neural network; Neural networks; Neurons; Pattern classification; Search methods;
Conference_Titel :
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5529-6
DOI :
10.1109/IJCNN.1999.832647
