Evolving artificial neural networks to control chaotic systems

by: Eric R Weeks, John M Burgess

Physical Review E, Vol. 56, No. 2. (1997), 1531.

View FullText article

DOI, PROLA (abstract), PROLA (pdf)

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

scis0000001, weeks

Find related articles with these CiteULike tags

chaos-control, misc, simulation

Abstract

We develop a genetic algorithm that produces neural network feedback controllers for chaotic systems. The algorithm was tested on the logistic and HÃ©non maps; for which it stabilizes an unstable fixed point using small perturbations; even in the presence of significant noise. The network training method [D. E. Moriarty and R. Miikkulainen; Mach. Learn. 22 ; 11 (1996)] requires no previous knowledge about the system to be controlled; including the dimensionality of the system and the location of unstable fixed points. This is the first dimension-independent algorithm that produces neural network controllers using time-series data. A software implementation of this algorithm is available via the World Wide Web.

@article{citeulike:849743, abstract = {We develop a genetic algorithm that produces neural network feedback controllers for chaotic systems. The algorithm was tested on the logistic and H\~{A}{\copyright}non maps; for which it stabilizes an unstable fixed point using small perturbations; even in the presence of significant noise. The network training method [D. E. Moriarty and R. Miikkulainen; Mach. Learn. 22 ; 11 (1996)] requires no previous knowledge about the system to be controlled; including the dimensionality of the system and the location of unstable fixed points. This is the first dimension-independent algorithm that produces neural network controllers using time-series data. A software implementation of this algorithm is available via the World Wide Web.}, author = {Weeks, Eric R. and Burgess, John M. }, citeulike-article-id = {849743}, doi = {10.1103/PhysRevE.56.1531}, journal = {Physical Review E}, keywords = {misc, simulation}, number = {2}, pages = {1531+}, posted-at = {2007-09-27 04:27:14}, priority = {0}, publisher = {American Physical Society}, title = {Evolving artificial neural networks to control chaotic systems}, url = {http://dx.doi.org/10.1103/PhysRevE.56.1531}, volume = {56}, year = {1997} }

RIS record

TY - JOUR ID - citeulike:849743 L3 - citeulike-article-id:849743 TI - Evolving artificial neural networks to control chaotic systems JF - Physical Review E VL - 56 IS - 2 SP - 1531 PB - American Physical Society N2 - We develop a genetic algorithm that produces neural network feedback controllers for chaotic systems. The algorithm was tested on the logistic and HÃ©non maps; for which it stabilizes an unstable fixed point using small perturbations; even in the presence of significant noise. The network training method [D. E. Moriarty and R. Miikkulainen; Mach. Learn. 22 ; 11 (1996)] requires no previous knowledge about the system to be controlled; including the dimensionality of the system and the location of unstable fixed points. This is the first dimension-independent algorithm that produces neural network controllers using time-series data. A software implementation of this algorithm is available via the World Wide Web. KW - misc KW - simulation AU - Weeks, Eric AU - Burgess, John PY - 1997/// UR - http://dx.doi.org/10.1103/PhysRevE.56.1531 ER -

RIS BibTeX