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.
Description
Informational Neurobayesian Approach to Neural Networks Training. Opportunities and Prospects
%0 Journal Article
%1 artemov2017informational
%A Artemov, Artem
%A Lutsenko, Eugeny
%A Ayunts, Edward
%A Bolokhov, Ivan
%D 2017
%J Arxiv.org
%K learning machine myown networks neural one-shot
%T Informational Neurobayesian Approach to Neural Networks Training. Opportunities and Prospects
%U http://arxiv.org/abs/1710.07264
%X 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.
@article{artemov2017informational,
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.},
added-at = {2017-10-26T16:40:38.000+0200},
author = {Artemov, Artem and Lutsenko, Eugeny and Ayunts, Edward and Bolokhov, Ivan},
biburl = {https://www.bibsonomy.org/bibtex/2fce96d6c9ee6f173224f3f17139d58f6/cogsys},
description = {Informational Neurobayesian Approach to Neural Networks Training. Opportunities and Prospects},
interhash = {a82dfe232394f3635b2f11ba968258e0},
intrahash = {fce96d6c9ee6f173224f3f17139d58f6},
journal = {Arxiv.org},
keywords = {learning machine myown networks neural one-shot},
note = {cite arxiv:1710.07264 9 pages, 5 figures, 2 tables; corrected typos},
timestamp = {2017-10-26T16:40:38.000+0200},
title = {Informational Neurobayesian Approach to Neural Networks Training. Opportunities and Prospects},
url = {http://arxiv.org/abs/1710.07264},
year = 2017
}