An accelerated learning algorithm for multilayer perceptron networks

Autor: Alexander G. Parlos, Benito R. Fernandez, Wei K. Tsai, J. Muthusami, Amir F. Atiya
Rok vydání: 1994
Předmět:
Zdroj: IEEE Transactions on Neural Networks. 5:493-497
ISSN: 1045-9227
DOI: 10.1109/72.286921
Popis: An accelerated learning algorithm (ABP-adaptive back propagation) is proposed for the supervised training of multilayer perceptron networks. The learning algorithm is inspired from the principle of "forced dynamics" for the total error functional. The algorithm updates the weights in the direction of steepest descent, but with a learning rate a specific function of the error and of the error gradient norm. This specific form of this function is chosen such as to accelerate convergence. Furthermore, ABP introduces no additional "tuning" parameters found in variants of the backpropagation algorithm. Simulation results indicate a superior convergence speed for analog problems only, as compared to other competing methods, as well as reduced sensitivity to algorithm step size parameter variations. >
Databáze: OpenAIRE
Pro tento záznam nejsou dostupné žádné jednotky.