Feedforward networks training speed enhancement by optimal initialization of the synaptic coefficients

IEEE Transactions on Neural Networks

Volumn 12, Issue 2, 2001, Pages 430-434

  Yam, J Y F ^a   Chow, T W S ^a  

^a CITY UNIVERSITY OF HONG KONG   (Hong Kong)

Author keywords

[No Author keywords available]

Indexed keywords

ALGORITHMS; BENCHMARKING; COMPUTER SIMULATION; VECTOR QUANTIZATION;

OPTIMAL INITIALIZATION; SYNAPTIC COEFFICIENTS;

FEEDFORWARD NEURAL NETWORKS;

EID: 0035272407     PISSN: 10459227     EISSN: None     Source Type: Journal    
DOI: 10.1109/72.914538     Document Type: Article

References (13)

1. Fast learning in artificial neural systems: Multilayer perceptron training using optimal estimation
2. Determining initial weights of feedforward neural networks based on least squares method
3. A new method in determining the initial weights of feedforward neural networks
4. Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights
5. New approach to selection of initial values of weights in neural function approximation
6. Statistically controlled activation weight initialization (SCAWI)
7. Avoiding false local minima by proper initialization of connections
8. High-order and multilayer perceptron initialization
9. A learning algorithm for multilayered neural networks based on linear least squares problems
10. Training feedforward networks with the Marquardt algorithm
11. A study on initial value setting of the back-propagation learning algorithm