Computers and Operations Research
MLP in layer-wise form with applications to weight decay
Neural Computation
Approximation by neural networks with a bounded number of nodes at each level
Journal of Approximation Theory
Constructive approximate interpolation by neural networks
Journal of Computational and Applied Mathematics
Electromagnetic field identification using artificial neural networks
NN'07 Proceedings of the 8th Conference on 8th WSEAS International Conference on Neural Networks - Volume 8
A New Constructive Algorithm for Designing and Training Artificial Neural Networks
Neural Information Processing
Sales forecasting using extreme learning machine with applications in fashion retailing
Decision Support Systems
International Journal of Systems Science
Equivalent Relationship of Feedforward Neural Networks and Real-Time Face Detection System
Proceedings of the FIRA RoboWorld Congress 2009 on Advances in Robotics
Constructive approximate interpolation by neural networks
Journal of Computational and Applied Mathematics
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Letters: Fully complex extreme learning machine
Neurocomputing
Approximate interpolation by neural networks with the inverse multiquadric functions
ISICA'07 Proceedings of the 2nd international conference on Advances in computation and intelligence
Constructive approximation to multivariate function by decay RBF neural network
IEEE Transactions on Neural Networks
Artificial Intelligence Review
IEEE Transactions on Neural Networks
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part I
Neural network based modelling of environmental variables: A systematic approach
Mathematical and Computer Modelling: An International Journal
Displacement prediction model of landslide based on ensemble of extreme learning machine
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part IV
Supporting product design by anticipating the success chances of new value profiles
Computers in Industry
Computer Speech and Language
Generalized single-hidden layer feedforward networks
ISNN'13 Proceedings of the 10th international conference on Advances in Neural Networks - Volume Part I
| Hi-index | 0.00 |
Neural-network theorems state that only when there are infinitely many hidden units is a four-layered feedforward neural network equivalent to a three-layered feedforward neural network. In actual applications, however, the use of infinitely many hidden units is impractical. Therefore, studies should focus on the capabilities of a neural network with a finite number of hidden units, In this paper, a proof is given showing that a three-layered feedforward network with N-1 hidden units can give any N input-target relations exactly. Based on results of the proof, a four-layered network is constructed and is found to give any N input-target relations with a negligibly small error using only (N/2)+3 hidden units. This shows that a four-layered feedforward network is superior to a three-layered feedforward network in terms of the number of parameters needed for the training data