In this study we investigate a hybrid neural network architecture for modelling purposes. The proposed network is based on the multilayer perceptron (MLP) network. However, in addition to the usual hidden layers the first hidden layer is selected to be a centroid layer. Each unit in this new layer incorporates a centroid that is located somewhere in the input space. The output of these units is the Euclidean distance between the centroid and the input. The centroid layer clearly resembles the hidden layer of the radial basis function (RBF) networks. Therefore the centroid based multilayer perceptron (CMLP) networks can be regarded as a hybrid of MLP and RBF networks. The presented benchmark experiments show that the proposed hybrid architecture is able to combine the good properties of MLP and RBF networks resulting fast and efficient learning, and compact network structure.
[1]
Martin A. Riedmiller,et al.
A direct adaptive method for faster backpropagation learning: the RPROP algorithm
,
1993,
IEEE International Conference on Neural Networks.
[2]
John Moody,et al.
Fast Learning in Networks of Locally-Tuned Processing Units
,
1989,
Neural Computation.
[3]
Jukka Saarinen,et al.
Weight Initialization Techniques in Neural Network Systems and Their Application
,
1997
.
[4]
Christian Lebiere,et al.
The Cascade-Correlation Learning Architecture
,
1989,
NIPS.
[5]
Julius T. Tou,et al.
Pattern Recognition Principles
,
1974
.
[6]
Simon Haykin,et al.
Neural Networks: A Comprehensive Foundation
,
1998
.
[7]
R. Fletcher.
Practical Methods of Optimization
,
1988
.