Title :
Chaos in a Hebbian-type learning algorithm
Author :
Li, Chunguong ; Yu, Juebang ; Liao, Xiaofeng
Author_Institution :
Inst. of Electron. Syst., Univ. of Electron. Sci. & Technol. of China, Chengdu, China
fDate :
29 June-1 July 2002
Abstract :
In this paper, it is shown that a Hebbian-type unsupervised learning algorithm can exhibit chaotic behavior and strange attractors. The Lyapunov exponent and power spectra are presented to confirm the existence of chaotic motion. A chaos-periodic-chaos transition was also observed in this system.
Keywords :
Hebbian learning; Lyapunov methods; adaptive systems; bifurcation; chaos; stability; unsupervised learning; Hebbian neural learning schemes; Hebbian-type unsupervised learning algorithms; Lyapunov exponents; adaptive systems; bifurcations; chaos-periodic-chaos transitions; chaotic behavior; chaotic motion power spectra; constrained anti-Hebbian learning; dynamic system deterministic nonperiodic behavior; strange attractors; Adaptive control; Adaptive filters; Adaptive systems; Chaos; Educational institutions; Least squares approximation; Neurons; Nonlinear dynamical systems; Power engineering and energy; Vectors;
Conference_Titel :
Communications, Circuits and Systems and West Sino Expositions, IEEE 2002 International Conference on
Print_ISBN :
0-7803-7547-5
DOI :
10.1109/ICCCAS.2002.1178999
