Title :
Analysis of the fluctuating logistic mapping for a Boltzmann machine
Author :
Okuhara, Koji ; Osaki, Shunji
Author_Institution :
Dept. of Ind. & Syst. Eng., Hiroshima Univ., Japan
Abstract :
Discusses the behavior of a single unit in a Boltzmann machine. Differentiating the sigmoid function of each unit with respect to an input, the authors introduce the fluctuating logistic mapping by using the finite difference method. The result shows that chaos generation is observed under some conditions. The authors further discuss chaos and learning.
Keywords :
Boltzmann machines; chaos; finite difference methods; learning (artificial intelligence); neural nets; Boltzmann machine; chaos generation; finite difference method; fluctuating logistic mapping; learning; sigmoid function; Chaos; Finite difference methods; Fluctuations; Logistics; Machine learning; Systems engineering and theory; Temperature;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.714196
