Informational Neurobayesian Approach to Neural Networks Training. Opportunities and Prospects

TL;DR

This paper introduces the Informational Neurobayesian Approach (INA), a novel neural network training method based on information theory that requires less data and computation, showing promising results compared to traditional methods.

Contribution

The paper proposes the INA, a new training approach that reduces data and computational requirements in neural networks, advancing the field of information-theoretic classification.

Findings

INA requires less training data

INA reduces computational power needed

INA shows promising classification performance

Abstract

A study of the classification problem in context of information theory is presented in the paper. Current research in that field is focused on optimisation and bayesian approach. Although that gives satisfying results, they require a vast amount of data and computations to train on. Authors propose a new concept named Informational Neurobayesian Approach (INA), which allows to solve the same problems, but requires significantly less training data as well as computational power. Experiments were conducted to compare its performance with the traditional one and the results showed that capacity of the INA is quite promising.