Summary

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

2012

Session Number:C3L-B

Session:

Number:719

Multi-Layer Perceptron Decided Leaning Neurons by Regular Output Glias

Chihiro Ikuta,  Yoko Uwate,  Yoshifumi Nishio,  Guoan Yang,  

pp.719-722

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.1.719

PDF download (895.2KB)

A glia is nervous cell which is existing in a brain. This cell changes a Ca²⁺ concentration and this ion affects a neuron learning. In the biological system, when the glia does not increase the Ca²⁺ concentration, the neuron cannot increase the response. From these features, we propose a Multi-Layer Perceptron (MLP) decided learning neurons by regular output glias. The neurons are separated to some groups. Each group changes a learning term and a non-learning term. We consider that a performance of the MLP improves by having two terms. By two different simulations, we confirm a learning ability and a characteristics of the proposed MLP.

[1] P.G. Haydon, “Glia: Listening and Talking to the Synapse,” Nature Reviews Neuroscience, vol. 2, pp. 844-847, 2001.

[2] S. Ozawa, “Role of Glutamate Transporters in Excitatory Synapses in Cerebellar Purkinje Cells,” Brain and Nerve, vol. 59, pp. 669-676, 2007.

[3] N. Takata, T. Mishima, C. Hisatsune, T. Nagai, E. Ebisui, K. Mikoshiba, and H. Hirase, “Astrocyte Calcium Signaling Transforms Cholinergic Modulation to Cortical Plasticity In Vivo,” The Journal of Neurocience, vol. 31, pp. 18155-18165, Dec. 2011.

[4] D.E. Rumelhart, G.E. Hinton and R.J. Williams, “Learning Representations by Back-Propagating Errors,” Nature, vol. 323-9, pp. 533-536, 1986.