Bayesian decisions with differentially fed neural networks

Authors:
R. Manjunath;K. S. Gurumurthy
Affiliations:
Bangalore University, Dept of EC & CSE, UVCE, Bangalore, India;Bangalore University, Dept of EC & CSE, UVCE, Bangalore, India
Venue:
ICECS'03 Proceedings of the 2nd WSEAS International Conference on Electronics, Control and Signal Processing
Year:
2003

Citing 4
Cited 0

Simulated annealing and Boltzmann machines: a stochastic approach to combinatorial optimization and neural computing

Simulated annealing and Boltzmann machines: a stochastic approach to combinatorial optimization and neural computing
A practical Bayesian framework for backpropagation networks

Neural Computation
Information geometry of the EM and em algorithms for neural networks

Neural Networks
Bayesian Learning for Neural Networks

Bayesian Learning for Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Generally the neural networks employing Bayesian decision do not output one simple hypothesis, but a manifold of probability distributions. This throws out the bayes posterior coefficients as a large number of classifiers. Here a novel method based on differential feedback is explored to merge these classifiers. The experimental results confirm affine transportation of these classifiers. Also, it has been shown that the differentially fed Artificial Neural Networks (ANNs) learn in much the same way as Bayesian learning and are hence resistant to over fitting.