Publication details

Strictly modular probabilistic neural networks for pattern recognition

Journal Article

Grim Jiří, Just P., Pudil Pavel

serial: Neural Network World vol.13, 6 (2003), p. 599-615

research: CEZ:AV0Z1075907

project(s): GA402/01/0981, GA ČR

keywords: neural networks, distribution mixtures, pattern recognition

abstract (eng):

Considering the statistical pattern recognition we approximate the unknown class-conditional probability distributions by multivariate Bernoulli mixtures. We show that both the parameter optimization based on EM algorithm and the resulting Bayesian decision-making can be realized by a strictly modular probabilistic neural network. The autonomous adaptation of neurons includes only the locally available information. The properties of the sequential learning procedure are illustrated by numerical examples.

Cosati: 09K, 12B, 06D