Summary

Proceedings of the 2012 International Symposium on Nonlinear Theory and its Applications

2012

Session Number:C2L-B

Session:

Number:644

Dependence of Sensitive Responses of Chaotic Wandering States on Configuration of Inhibited In-Coming Synaptic Connections

Toshiyuki Hamada,  Jousuke Kuroiwa,  Hisakazu Ogura,  Tomohiro Odaka,  Izumi Suwa,  Haruhiko Shirai,  

pp.644-647

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.1.644

PDF download (1.7MB)

In this paper, we investigate the dependence of the sensitive responses of the chaotic wandering states to memory pattern fragments on the configuration of the inhibited in-coming synaptic connections in Nara & Davis chaotic neural network model. It has been shown that Nara & Davis model can quickly access to the target pattern corresponding to the input memory pattern fragment with almost 3 steps and perfect possibility. However, the potentiality depends on the dynamical property of the chaotic wandering states, which changes with the configuration of the inhibited in-coming synaptic connections. Therefore, the purpose of this paper is to investigate the difference of the sensitivity to memory pattern fragments among the configurations of the inhibited in-coming synaptic connections. From computer experiments, with the configuration which does not include in-coming synaptic connections from the elements of memory pattern fragments, the model has quite higher potentiality in the accessing time and the success ratio than the configuration which includes the connections. In addition, from the viewpoints of Lyapunov dimension and information dimension, the model without in-coming synaptic connections from the fragments reveals higher developed chaos.

[1] SA. Skarda, J. K. Freeman, “How brains make chaos in order to make sense of the world?” Behavioral and Brain Sci., vol.10, pp.161-195, 1987

[2] S. Mikami, S. Nara “Dynamical response of chaotic memory dynamics to weak input in a recurrent neural network model”, Neural Comput. Appl. 11, pp.129-136, 2003

[3] J. Kuroiwa, S. Masutani, S. Nara, K. Aihara, “Sensitive responses of chaotic wandering to memory pattern fragments inputs in a chaotic neural network model.” Int. J. Bifurcation & Chaos, vol.14, pp.1413-1421, 2004

[4] T. Hamada, J. Kuroiwa, H. Ogura, T. Odaka, H. Shirai, Y. Kato, “Response Properties to Inputs of Memory Pattern Fragments in Three Types of Chaotic Neural Network Models”, In: Proceeding of 19th International Conference on Articial Neural Networks (ICANN2009) pp.544-551, 2009

[5] T. Hamada, J. Kuroiwa, H. Ogura, T. Odaka, H. Shirai, I. Suwa, “Dependence on Memory Pattern in Sensitive Response of Memory Pattern Fragments among Three Types of Chaotic Neural Network Models”, In: Proceeding of 17th International Conference on Neural Information Processing (ICONIP2010), pp.223-230, 2010