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: |
Wake-sleep algorithm
Adaptive algorithm Artificial neural network Computer Networks and Communications Computer science Population-based incremental learning Supervised learning Stability (learning theory) General Medicine Perceptron Generalization error Backpropagation Computer Science Applications Artificial Intelligence Multilayer perceptron Gradient descent Algorithm Software |
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 |
Externí odkaz: |
Pro tento záznam nejsou dostupné žádné jednotky.