Piecewise-linear classifiers, formal neurons and separability of the learning sets

The design of piecewise-linear classifiers from formal neurons is considered. The design classifiers are based on hierarchical, multilayer neural networks. The described procedure allows to find both the structure of network (the numbers of layers and neurons) and weights of single neurons. The main...

Full description

Saved in:
Bibliographic Details
Published inProceedings of 13th International Conference on Pattern Recognition Vol. 4; pp. 224 - 228 vol.4
Main Author Bobrowski, L.
Format Conference Proceeding
LanguageEnglish
Published IEEE 1996
Subjects
Biomedical engineering
Computational modeling
Computer networks
Computer science
Multi-layer neural network
Network synthesis
Neural networks
Neurons
Pattern recognition
Piecewise linear techniques
Online AccessGet full text
ISBN9780818672828
081867282X
ISSN1051-4651
DOI10.1109/ICPR.1996.547420

Cover

More Information
Summary:The design of piecewise-linear classifiers from formal neurons is considered. The design classifiers are based on hierarchical, multilayer neural networks. The described procedure allows to find both the structure of network (the numbers of layers and neurons) and weights of single neurons. The main principle of the synthesis procedure is to preserve separability of learning sets during data compression by successive neural layers. Different procedures aiming at improving the network compression ability are also considered.
ISBN:9780818672828
081867282X
ISSN:1051-4651
DOI:10.1109/ICPR.1996.547420