A note on activation function in multilayer feedforward learning

Proceedings of the International Joint Conference on Neural Networks

Volumn 1, Issue , 2002, Pages 519-523

  Kamruzzaman, Joarder ^a   Aziz, S M ^a  

^a UNIVERSITY OF BALLARAT   (Australia)

Author keywords

Activation function; Backpropagation learning; Learning speed; Neural network

Indexed keywords

BACKPROPAGATION; COMPUTER SIMULATION; CONVERGENCE OF NUMERICAL METHODS; FEEDFORWARD NEURAL NETWORKS; LEARNING ALGORITHMS; MATHEMATICAL MODELS; VECTORS;

ACTIVATION FUNCTION; MULTILAYER FEEDFORWARD LEARNING;

MULTILAYER NEURAL NETWORKS;

EID: 0036079318     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (14)

1. Comparison of generalization multiplayer perceptron with log-likelihood and least squares cost function
2. Supervised learning using the cauchy energy function
3. Improving the convergence of the back-propagation algorithm
4. Increased rates of convergence through learning rate adaptation
5. A method for self determination of adaptive learning rates in back propagation
6. Acceleration of backpropagation learning using optimised learning rate and momentum
7. Fast training of multilayered feedforward neural networks
8. Fast training of multilayered perceptrons
9. Fast convergent generalized backpropagation algorithm with constant learning rate
10. The backpropagation learning with logarithmic transfer function
11. An empirical study of learning speed in backpropagation networks
12. A direct adaptive method for faster backpropagation learning: The EPROP algorithm