11814 search results - page 70 / 2363 » On Generalization by Neural Networks |
125
Voted
ECML
2007
Springer
15 years 10 months ago
2007
Springer
Abstract. The paper investigates modification of backpropagation algorithm, consisting of discretization of neural network weights after each training cycle. This modification, a...
135
Voted
GECCO
2007
Springer
15 years 10 months ago
2007
Springer
A significant problem for evolving artificial neural networks is that the physical arrangement of sensors and effectors is invisible to the evolutionary algorithm. For example,...
135
Voted
FLAIRS
2006
15 years 5 months ago
2006
Fuzzy ARTMAP (FAM) is one of the best neural network architectures in solving classification problems. One of the limitations of Fuzzy ARTMAP that has been extensively reported in...
127
click to vote
CORR
2010
Springer
15 years 3 months ago
2010
Springer
Although backpropagation ANNs generally predict better than decision trees do for pattern classification problems, they are often regarded as black boxes, i.e., their predictions c...
140
Voted
CDC
2008
IEEE
15 years 10 months ago
2008
IEEE
— This paper studies the problem of stability analysis for neural networks (NNs) with a time-varying delay. The activation functions are assumed to be neither monotonic, nor diff...