2924 search results - page 146 / 585 » On the Use of Evidence in Neural Networks |
113
Voted
ISNN
2007
Springer
15 years 8 months ago
2007
Springer
Normal fuzzy CMAC neural network performs well because of its fast learning speed and local generalization capability for approximating nonlinear functions. However, it requires hu...
104
Voted
ICANN
2005
Springer
15 years 7 months ago
2005
Springer
This work proposes a theoretical guideline in the specific area of Frequent Itemset Mining (FIM). It supports the hypothesis that the use of neural network technology for the prob...
146
click to vote
EGITALY
2006
15 years 3 months ago
2006
Palette re-ordering is a well known and very effective approach for improving the compression of color-indexed images. If the spatial distribution of the indexes in the image is s...
110
Voted
ICANN
2007
Springer
15 years 8 months ago
2007
Springer
Abstract. Reservoir Computing (RC) uses a randomly created recurrent neural network where only a linear readout layer is trained. In this work, RC is used for detecting complex eve...
105
click to vote
CONNECTION
2004
15 years 1 months ago
2004
: High capacity associative neural networks can be built from networks of perceptrons, trained using simple perceptron training. Such networks perform much better than those traine...