Original Contribution: A quantitative analysis of the behaviors of the PLN network

Authors:
Bo Zhang;Ling Zhang;Huai Zhang
Affiliations:
Tsinghua University, China;Anqing Teachers' College, China;Tsinghua University, China
Venue:
Neural Networks
Year:
1992

Citing 2
Cited 1

The logic of connectionist systems

Neural computing architectures
A probabilistic logic neuron network for associative learning

Neural computing architectures

On a learnability question associated to neural networks with continuous activations (extended abstract)

COLT '94 Proceedings of the seventh annual conference on Computational learning theory

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, the main behavior of PLN networks with feedback connections are investigated quantitatively by using Markov chains theory. The main results obtained are (a) the conditions that a PLN network converges, (b) the probability that each state in a given network converges to a stable state, (c) the mean and the variance of the number of steps that each state in a given network converges to a set of stable states, and (d) the upper bound of the average number of steps that each state in an unknown PLN network converges. A number of computer simulations are given to verify the analysis. These results not only give an insight to the details of the behavior of PLN network, but also show that the network is a promising model for connectionist system.