The optimised internal representation of multilayer classifier networks performs nonlinear discriminant analysis

Neural Networks

Volumn 3, Issue 4, 1990, Pages 367-375

  Webb, Andrew R ^a   Lowe, David ^a  

^a DEFENCE SCIENCE AND TECHNOLOGY LABORATORY   (United Kingdom)

Author keywords

Adaptive layered networks; Learning; Nonlinear discriminant analysis; Nonlinear optimisation; Pattern classification

Indexed keywords

OPTIMIZATION; PATTERN RECOGNITION;

ADAPTIVE LAYERED NETWORKS; FEEDFORWARD NETWORKS; NONLINEAR DISCRIMINANT ANALYSIS; PATTERN CLASSIFICATION;

NEURAL NETWORKS;

EID: 0025208603     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/0893-6080(90)90019-H     Document Type: Article

References (22)

1. Multi-variable functional interpolation and adaptive networks
2. Quasi-Newton methods and their application to function minimisation
3. Large automatic learning, rule extraction, and generalisation
4. Multilayer perceptrons and data analysis
5. Calculating the singular values and pseudo-inverse of a matrix
6. Analysis of hidden units in a layered network trained to classify sonar targets
7. Disjunctive models of Boolean category learning
8. Knowledge representation in connectionist networks
9. Neural net and traditional classifiers
10. Adaptive training of multilayer neural networks using a least squares estimation technique
11. An introduction to computing with neural nets
12. Adaptive radial basis function nonlinearities and the problem of generalisation
13. A neural model for category learning
14. Learning system architectures composed of multiple learning modules
15. Learning algorithms for connectionist networks: Applied gradient methods of nonlinear optimisation
16. A comparison of nonlinear optimisation strategies for feed-forward adaptive layered networks
17. A hybrid optimisation strategy for adaptive feed-forward layered networks