Sciweavers

2924 search results - page 44 / 585
» On the Use of Evidence in Neural Networks
Sort
View
IJCNN
2006
IEEE
15 years 7 months ago
Ensemble Techniques for Avoiding Poor Performance in Evolved Neural Networks
— The idea of using evolutionary techniques to optimize the performance of neural networks is now widely used, but some approaches have been found to result in the evolution of r...
John A. Bullinaria
ISCI
2002
120views more  ISCI 2002»
15 years 1 months ago
A class of instantaneously trained neural networks
This paper presents FC networks that are instantaneously trained neural networks that allow rapid learning of non-binary data. These networks, which generalize the earlier CC netw...
Subhash C. Kak
NN
2002
Springer
113views Neural Networks» more  NN 2002»
15 years 1 months ago
A dynamically coupled neural oscillator network for image segmentation
We propose a dynamically coupled neural oscillator network for image segmentation. Instead of pair-wise coupling, an ensemble of oscillators coupled in a local region is used for ...
Ke Chen 0001, DeLiang L. Wang
ISDA
2006
IEEE
15 years 7 months ago
Modular Neural Network Task Decomposition Via Entropic Clustering
The use of monolithic neural networks (such as a multilayer perceptron) has some drawbacks: e.g. slow learning, weight coupling, the black box effect. These can be alleviated by t...
Jorge M. Santos, Luís A. Alexandre, Joaquim...
FUZZY
2001
Springer
184views Fuzzy Logic» more  FUZZY 2001»
15 years 6 months ago
Handwritten Digit Recognition: A Neural Network Demo
Abstract. A handwritten digit recognition system was used in a demonstration project to visualize artificial neural networks, in particular Kohonen’s self-organizing feature map...
Berend-Jan van der Zwaag