A Polynomial Time Algorithm for Generating Neural Networks for Pattern Classification: Its Stability Properties and Some Test Results
- 1 March 1993
- journal article
- Published by MIT Press in Neural Computation
- Vol. 5 (2) , 317-330
- https://doi.org/10.1162/neco.1993.5.2.317
Abstract
Polynomial time training and network design are two major issues for the neural network community. A new algorithm has been developed that can learn in polynomial time and also design an appropriate network. The algorithm is for classification problems and uses linear programing models to design and train the network. This paper summarizes the new algorithm, proves its stability properties, and provides some computational results to demonstrate its potential.Keywords
This publication has 5 references indexed in Scilit:
- Pattern Classification Using Linear ProgrammingINFORMS Journal on Computing, 1991
- Improved Linear Programming Models for Discriminant Analysis*Decision Sciences, 1990
- Limitations of multi-layer perceptron networks - steps towards genetic neural networksParallel Computing, 1990
- A new polynomial-time algorithm for linear programmingCombinatorica, 1984
- A neural model for category learningBiological Cybernetics, 1982