Dynamic stability conditions for Lotka-Volterra recurrent neural networks with delays
DOI: 10.1103/PHYSREVE.66.011910
关键词:
摘要: The Lotka-Volterra model of neural networks, derived from the membrane dynamics competing neurons, have found successful applications in many "winner-take-all" types problems. This paper studies dynamic stability properties general recurrent networks with delays. Conditions for nondivergence are derived. These conditions based on local inhibition thereby allowing these to possess a multistability property. Multistability is necessary property network that will enable important computations such as those governing decision making process. Under conditions, compact set globally attracts all trajectories can be computed explicitly. If connection weight matrix symmetric some sense, and delays L2 space, we prove complete stability.
harvard.edu
sci-hub.se 参考文章(24)
Tomoki Fukai, Shigeru Tanaka, A simple neural network exhibiting selective activation of neuronal ensembles: from winner-take-all to winners-share-all Neural Computation. ,vol. 9, pp. 77- 97 ,(1997) , 10.1162/NECO.1997.9.1.77
T. Asai, T. Fukai, S. Tanaka, A subthreshold MOS circuit for the Lotka-Volterra neural network producing the winners-share-all solution Neural Networks. ,vol. 12, pp. 211- 216 ,(1999) , 10.1016/S0893-6080(98)00121-X
T. Asai, M. Ohtani, H. Yonezu, Analog integrated circuits for the Lotka-Volterra competitive neural networks IEEE Transactions on Neural Networks. ,vol. 10, pp. 1222- 1231 ,(1999) , 10.1109/72.788661
J. J. Hopfield, Neurons with graded response have collective computational properties like those of two-state neurons. Proceedings of the National Academy of Sciences of the United States of America. ,vol. 81, pp. 3088- 3092 ,(1984) , 10.1073/PNAS.81.10.3088
P. Baldi, A.F. Atiya, How delays affect neural dynamics and learning IEEE Transactions on Neural Networks. ,vol. 5, pp. 612- 621 ,(1994) , 10.1109/72.298231
C. M. Marcus, R. M. Westervelt, Stability of analog neural networks with delay Physical Review A. ,vol. 39, pp. 347- 359 ,(1989) , 10.1103/PHYSREVA.39.347
P.L. Venetianter, T. Roska, Image compression by cellular neural networks IEEE Transactions on Circuits and Systems I-regular Papers. ,vol. 45, pp. 205- 215 ,(1998) , 10.1109/81.662694
S. Arik, Stability analysis of delayed neural networks IEEE Transactions on Circuits and Systems I-regular Papers. ,vol. 47, pp. 1089- 1092 ,(2000) , 10.1109/81.855465
P. van den Driessche, Xingfu Zou, Global Attractivity in Delayed Hopfield Neural Network Models SIAM Journal on Applied Mathematics. ,vol. 58, pp. 1878- 1890 ,(1998) , 10.1137/S0036139997321219
M. Gilli, Stability of cellular neural networks and delayed cellular neural networks with nonpositive templates and nonmonotonic output functions IEEE Transactions on Circuits and Systems I-regular Papers. ,vol. 41, pp. 518- 528 ,(1994) , 10.1109/81.311541