Title :
Iterative generation of higher-order nets in polynomial time using linear programming
Author :
Roy, Asim ; Mukhopadhyay, Somnath
Author_Institution :
Dept. of Decision & Inf. Syst., Arizona State Univ., Tempe, AZ, USA
fDate :
3/1/1997 12:00:00 AM
Abstract :
This paper presents an algorithm for constructing and training a class of higher-order perceptrons for classification problems. The method uses linear programming models to construct and train the net. Its polynomial time complexity is proven and computational results are provided for several well-known problems. In all cases, very small nets were created compared to those reported in other computational studies
Keywords :
computational complexity; feedforward neural nets; iterative methods; learning (artificial intelligence); linear programming; multilayer perceptrons; pattern classification; feedforward neural nets; higher-order neural nets; learning; linear programming; multilayer perceptrons; pattern classification; polynomial time complexity; Biological neural networks; Feedforward neural networks; Iterative algorithms; Learning systems; Linear programming; Mathematics; Multi-layer neural network; Multilayer perceptrons; Neural networks; Polynomials;
Journal_Title :
Neural Networks, IEEE Transactions on
