Stability of Periodic Orbits and Fault Tolerance in Dynamic Binary Neural Networks

The dynamic binary neural network is characterized by ternary connection parameters and can generate various binary periodic orbits. This paper considers two interesting problems based on typical examples. First, effect of connection sparsity on stability of target periodic orbits is considered: adding branches adequately to the most sparse network, stability of the periodic orbits can be reinforced. Second, fault tolerance of the network is considered: cutting one branch from the network, storage and stability of the periodic orbits are preserved in high probability.

[1]  Yoshua Bengio,et al.  BinaryConnect: Training Deep Neural Networks with binary weights during propagations , 2015, NIPS.

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

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

[4]  Toshimichi Saito,et al.  Stabilization of desired periodic orbits in dynamic binary neural networks , 2017, Neurocomputing.

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

[6]  Toshimichi Saito,et al.  Connection sparsity versus orbit stability in dynamic binary neural networks , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[7]  M. Lavanya,et al.  Design and Analysis of Linear Feedback Shift Register Based on Various Tap Connections , 2012 .

[8]  Sung-Kwon Park,et al.  The geometrical learning of binary neural networks , 1995, IEEE Trans. Neural Networks.

[9]  Vincent Gripon,et al.  Storing Sequences in Binary Tournament-Based Neural Networks , 2014, IEEE Transactions on Neural Networks and Learning Systems.

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