1. Strictly modular probabilistic neural networks for pattern recognition
- Creator:
- Grim, Jiří, Just, Petr, and Pudil, Pavel
- Format:
- bez média and svazek
- Type:
- model:article and TEXT
- Subject:
- probabilistic neural networks, distribution mixtures, strictly modular properties, and sequential EM algorithm
- Language:
- English
- Description:
- Considering the statistical recognition of multidimensional binary observations we approximato the unknown class-conditioiial probability distributions by multivariate Bernoulli mixtures. We show that both the parameter optimization and the resulting Bayesian decision-making can be realized by a probabilistic neural network having strictly modular properties. In particular, the process of learning based on the EM algorithm can be perfomied by means of a sequential autonomous adaptation of neurons involving only the infomiation from the input synapses and the interior of neurons. In this sense the probabilistic neural network can be designed automatically. The properties of the sequential strictly modular learning procedure are illustrated by mumerical exainples.
- Rights:
- http://creativecommons.org/publicdomain/mark/1.0/ and policy:public