A continuous attractor of a recurrent neural network is a set of connected stable equilibrium points. The continuous attractors of neural networks with symmetric connection weights have been studied widely. However, most of the matrixes are asymmetric. In this paper, the continuous attractors of 3-D discrete-time ring networks with circulant weight matrix are studied. The circulant matrix is asymmetric. The eigenvalues of asymmetric matrix may be complex. Based on the complex eigenvalues, the conditions that guarantee the networks with circulant weight matrix to have continuous attractors are obtained.
CITATION STYLE
Yu, J., Yi, Z., Liao, Y., Wu, D. A., & Dai, X. (2018). Continuous attractors of 3-D discrete-time ring networks with circulant weight matrix. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10878 LNCS, pp. 388–396). Springer Verlag. https://doi.org/10.1007/978-3-319-92537-0_45
Mendeley helps you to discover research relevant for your work.