A fast fixed point learning method to implement associative memory on CNNs

Cellular Neural Networks (CNNs) with space-varying interconnections are considered here to implement associative memories. A fast learning method is presented to compute the interconnection weights. The algorithm was carefully tested and compared to other methods. Storage capacity, noise immunity, and spurious state avoidance capability of the proposed system are discussed.

[1]  Josef A. Nossek,et al.  Design and learning with cellular neural networks , 1994, Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94).

[2]  M. Hasler,et al.  Information processing using stable and unstable oscillations: a tutorial , 1994, Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94).

[3]  Lin-Bao Yang,et al.  Cellular neural networks: theory , 1988 .

[4]  Iztok Fajfar,et al.  Finding all stable equilibria of cellular neural networks , 1994, Int. J. Circuit Theory Appl..

[5]  Tamás Roska,et al.  The CNN universal machine: an analogic array computer , 1993 .

[6]  Derong Liu,et al.  Cellular neural networks for associative memories , 1993 .

[7]  Joos Vandewalle,et al.  Cellular neural networks as a model of associative memories , 1990, IEEE International Workshop on Cellular Neural Networks and their Applications.

[8]  Leon O. Chua,et al.  The CNN paradigm , 1993 .