利用訊號連接的想法,重新分析關聯記憶的一種動力函數.原有的記憶訊息為了回覆記憶,採用Hebb的規則形成權值矩陣這些記憶訊息在分類後形成數個網路,而這些網路的結構就決定了動力函數的疊代收斂行為.我們可以知道這個動力函數回覆記憶的極限與限制.最後為了連結網路而產生更強大的網路,我們再增加一個新的記憶訊息來連結它們.

The Content-Addressable Memory Problem is one of the most famous in neural nets. This problem is recently settled by Shih-Tsai. In this thesis, we introduce the notion of the connection and give a different approach to some of results obtained in Shih-Tsai. We shall show that the nets will govern the fixed points base, the transient length, and the behavior of the dynamic equation.

[1] Konstantinos I. Diamantaras (2002), Neural Networks and Principal Component Analysis, CRC Press LLC.
[2] Yu Hen Hu and Jenq-Neng Hwang (2002), Introduction to Neural Networks for Signal Processing, CRC Press LLC.
[3] Hopfield, J. J. (1982), Neural networks and physical systems with emergent collective computational abilities, Proc Natl Acad Sci USA, vol 79, 2554-2558.
[4] Anil K. Jain, Jianchang Mao and K.M. Mohiuddin (1996), Artificial Neural Networks: A Tutorial. Computer, Mar.1996: 31-44.
[5] MemoryYves Kamp and Martin Hasler (1990), Recursive Neural Networks for Associative, John Wiley & Sons.
[6] Ben Kröse and Patrick van der Smagt (1996), An introduction to Neural Networks, Amsterdam University Press.
[7] Nikola K. Kasabov (1996), Foundations of Neural Networks, Fussy Systems, and Knowledge Engineering, Massachusetts Institute of Technology.
[8] Jose C. Principe (2000), Artificial Neural Networks, CRC Press LLC.
[9] Jose C. Principe (2002), Dynamic Neural Networks and Optimal Signal Processing, CRC Press LLC.
[10] M.-H. Shih and F.S. Tsai (2003), Content-addressable memory problem.
[11] McCulloch, W. and Pitts, W. (1943), A logical calculus of the ideas immanent in nervous activity, Bull Math Biophys 5:115-133.
[12] Gérard Weisbuch (1991), Complex Systems Dynamics, Addison-Wesley Publishing Company.