Basic Analysis of Cellular Dynamic Binary Neural Networks

This paper studies cellular dynamic binary neural networks that can generate various periodic orbits. The networks is characterized by signum activation function and local connection parameters. In order to visualize/analyze the dynamics, we present a feature plane of present two simple feature quantities. We also we present normal form equations that can describe all dynamics of the networks. Using the normal form equation and feature plane, various phenomena are investigated.

[1]  Paul L. Rosin.  Training cellular automata for image processing , 2005, IEEE Transactions on Image Processing.

[2]  Anthony N. Michel,et al.  A training algorithm for binary feedforward neural networks , 1992, IEEE Trans. Neural Networks.

[3]  Toshimichi Saito,et al.  Learning of Simple Dynamic Binary Neural Networks , 2013, IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences.

[4]  Toshimichi Saito,et al.  Stability of Periodic Orbits in Dynamic Binary Neural Networks with Ternary Connection , 2016, ICONIP.

[5]  Toshimichi Saito,et al.  Analysis of various periodic orbits in simple dynamic binary neural networks , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[6]  Guanrong Chen,et al.  Universal Perceptron and DNA-Like Learning Algorithm for Binary Neural Networks: Non-LSBF Implementation , 2009, IEEE Transactions on Neural Networks.

[7]  Hiroyuki Torikai,et al.  Theoretical and Heuristic Synthesis of Digital Spiking Neurons for Spike-Pattern-Division Multiplexing , 2010, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[8]  E. Ott.  Chaos in Dynamical Systems: Contents , 1993 .