具有轴突信号传输延迟的双向联想记忆神经网络
NEURAL NETWORKS OF BIDIRCTIONAL ASSOCIATIVE MEMORY WITH AXONAL SIGNAL TRANSMISSION DELAYS
-
摘要: 双向联想记忆模型是两层异联想网络,本文重点讨论了具有轴突信号传输延迟的双向联想记忆神经网络平衡点存在的条件以及全局稳定性.并用具体例子验证了结论的正确性。本文所讨论的方法具有普遍性,所得到的结果对于网络的设计和应用具有重要的指导意义。Abstract: Bidirectional Associative Memory(BAM) models are two-layer heteroassciative networks. In this paper, conditions for the existence of equilibrium points and global stability are emphatically disscussed for BAM with axonal signal transmission delays. The correctness of obtained conclusions is verified by use of some examples. The disscussed methods are more general. The obtained results have important leading significance in the design and application of BAM.
-
Kosko B. Bidirectional associative memories. IEEE Trans on SMC. 1988, SMC-18(1): 49-60.[2]Kosko B. Adaptive bidirectional associative memories. Appl. Opt. 1987, 26(23): 4947-4960.[3]Kosko B. Unsuperised learning in noise. IEEE Trans on NN, 1990, 1(1): 44-57.[4]Kosko B. Structural stability of unsuperised learning in feedback neural networks. IEEE Trans on AC, 1991, AC-36(7): 785-792.[5]Kosko B. Neural Networks and Fuzzy Systems--A Dynamical System Appreach to Machine Intelligence. Prentice Hall International Inc, 1992, 2-3: 38-108.[6]Mathai C.[J].Vpadhyaya B C. Performance Analysis and Application of the Bidirectional Aasocative Memory to Industrial Spectral Signatures. Proc. IJCNN 1989, Vol.l, Washington: June.1989,:-[7]廖晓峰,刘光远,虞厥邦.连续BAM网络的定性分析.电路与系统学报,1996, 1(2): 13-18.[8]Kohonen T. Self-Organzation and Associative Memery. New York: Springet Verlag, 1988.[9]Hirsch M W, Smale S. Differential Equations, Dynamical Systems and Linear Algebra: Orlando, FL: Acadmic press, 1974.[10]Michel A N, Miller R K. Qualitative Analysis of Large Scale Dynamical Systems. New York: Academic, 1977. -
计量
- 文章访问数: 3034
- HTML全文浏览量: 166
- PDF下载量: 362
- 被引次数: 0
下载:
