The position and stability of metastable states in asymmetric feedback neural networks
FANG Yi-bo, JIN Tao, QU Shi-xian *
(College of Physics and Information Technology, Shaanxi Normal University, Xi′an 710062, Shaanxi, China)
Abstract:
The positions of metastable states of single-layer feedback neural network designed by the generalized perceptron rule are investigated in the phase space. The Hamming distance between metastable states and ground states is calculated, which shows that the metastable states concentrate in the positions where approximately equidistant to all the ground states. In addition, the energy ratio of metastable states and ground states are calculated, which can be found that the influence of metastable states can be weakened by adjusting the parameter κ.
KeyWords:
neural network; metastable state; phase space; stability
