Building a Chaotic Proved Neural Network
Artificial Intelligence
2015-03-17 v1 Cryptography and Security
Dynamical Systems
General Topology
Abstract
Chaotic neural networks have received a great deal of attention these last years. In this paper we establish a precise correspondence between the so-called chaotic iterations and a particular class of artificial neural networks: global recurrent multi-layer perceptrons. We show formally that it is possible to make these iterations behave chaotically, as defined by Devaney, and thus we obtain the first neural networks proven chaotic. Several neural networks with different architectures are trained to exhibit a chaotical behavior.
Cite
@article{arxiv.1101.4351,
title = {Building a Chaotic Proved Neural Network},
author = {Jacques M. Bahi and Christophe Guyeux and Michel Salomon},
journal= {arXiv preprint arXiv:1101.4351},
year = {2015}
}
Comments
6 pages, submitted to ICCANS 2011