36 search results - page 1 / 8 » Taxonomy of Neural Transfer Functions |
IJCNN
2000
IEEE
13 years 9 months ago
2000
IEEE
The choice of transfer functions may strongly influence complexity and performance of neural networks used in classification and approximation tasks. A taxonomy of activation an...
ESANN
2001
13 years 6 months ago
2001
Abstract. Sigmoidal or radial transfer functions do not guarantee the best generalization nor fast learning of neural networks. Families of parameterized transfer functions provide...
IJCNN
2008
IEEE
13 years 11 months ago
2008
IEEE
— Numerical condition affects the learning speed and accuracy of most artificial neural network learning algorithms. In this paper, we examine the influence of opposite transfe...
ESANN
2001
13 years 6 months ago
2001
Neural networks use neurons of the same type in each layer but such architecture cannot lead to data models of optimal complexity and accuracy. Networks with architectures (number ...
FOCI
2007
IEEE
13 years 11 months ago
2007
IEEE
— Backpropagation through time is a very popular discrete-time recurrent neural network training algorithm. However, the computational time associated with the learning process t...