Title :
An extended RTRL training algorithm using Hessian matrix
Author :
Coelho, Pedro H G
Author_Institution :
State Univ. of Rio de Janeiro, Brazil
Abstract :
An extended real time recurrent learning (RTRL) algorithm using Hessian matrix is proposed. The algorithm is suitable for small fully recurrent neural networks present in several applications. Simulation results indicate that the training algorithm is fast
Keywords :
Hessian matrices; Newton method; convergence of numerical methods; learning (artificial intelligence); real-time systems; recurrent neural nets; transfer functions; Hessian matrix; Newton method; convergence; real time recurrent learning; recurrent neural networks; Adaptive control; Concurrent computing; Feedforward neural networks; Hardware; Information processing; Neural networks; Neurons; Newton method; Recurrent neural networks; Signal processing algorithms;
Conference_Titel :
Neural Networks, 2000. IJCNN 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on
Conference_Location :
Como
Print_ISBN :
0-7695-0619-4
DOI :
10.1109/IJCNN.2000.860831
