Title :
An algorithm for fast convergence in training neural networks
Author :
Wilamowski, Bogdan M. ; Iplikci, Serdar ; Kaynak, Okyay ; Efe, M. Onder
Author_Institution :
Graduate Center, Idaho Univ., Boise, ID, USA
Abstract :
In this work, two modifications on Levenberg-Marquardt (LM) algorithm for feedforward neural networks are studied. One modification is made on performance index, while the other one is on calculating gradient information. The modified algorithm gives a better convergence rate compared to the standard LM method and is less computationally intensive and requires less memory. The performance of the algorithm has been checked on several example problems
Keywords :
Jacobian matrices; convergence; feedforward neural nets; learning (artificial intelligence); performance index; Jacobian matrix; Levenberg-Marquardt algorithm; convergence rate; feedforward neural networks; gradient information; learning; performance index; Backpropagation algorithms; Convergence; Equations; Feedforward neural networks; Intelligent networks; Jacobian matrices; Neural networks; Newton method; Performance analysis; Stochastic processes;
Conference_Titel :
Neural Networks, 2001. Proceedings. IJCNN '01. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-7044-9
DOI :
10.1109/IJCNN.2001.938431
