A Boolean Hebb rule for binary associative memory design

A binary associative memory design procedure that gives a Hopfield network with a symmetric binary weight matrix is introduced in this paper. The proposed method is based on introducing the memory vectors as maximal independent sets to an undirected graph, which is constructed by Boolean operations analogous to the conventional Hebb rule. The parameters of the resulting network is then determined via the adjacency matrix of this graph in order to rind a maximal independent set whose characteristic vector is close to the given distorted vector. We show that the method provides attractiveness for each memory vector and avoids spurious memories whenever the set of given memory vectors satisfy certain compatibility conditions, which implicitly imply sparsity. The applicability of the design method is finally investigated by a quantitative analysis of the compatibility conditions.

[1]  Arun Jagota,et al.  Approximating maximum clique with a Hopfield network , 1995, IEEE Trans. Neural Networks.

[2]  Mohamad H. Hassoun,et al.  A two-level Hamming network for high performance associative memory , 2001, Neural Networks.

[3]  Gottfried Tinhofer,et al.  A branch and bound algorithm for the maximum clique problem , 1990, ZOR Methods Model. Oper. Res..

[4]  Sudhakar M. Reddy,et al.  Guaranteed convergence in a class of Hopfield networks , 1992, IEEE Trans. Neural Networks.

[5]  A.N. Michel,et al.  Associative memories via artificial neural networks , 1990, IEEE Control Systems Magazine.

[6]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[7]  J. Moon,et al.  On cliques in graphs , 1965 .

[8]  L. Personnaz,et al.  Collective computational properties of neural networks: New learning mechanisms. , 1986, Physical review. A, General physics.

[9]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[10]  N. S. Şengör,et al.  An analysis of maximum clique formulations and saturated linear dynamical network , 1999 .

[11]  Zoltán Füredi,et al.  The number of maximal independent sets in connected graphs , 1987, J. Graph Theory.

[12]  Jacek M. Zurada,et al.  Introduction to artificial neural systems , 1992 .

[13]  Soura Dasgupta,et al.  Nonpositive Hopfield networks for unidirectional error correcting coding , 1995 .

[14]  A. Michel,et al.  Analysis and synthesis of a class of neural networks: linear systems operating on a closed hypercube , 1989 .

[15]  Ferhan Pekergin,et al.  A saturated linear dynamical network for approximating maximum clique , 1999 .

[16]  Paul Erdös,et al.  Clique numbers of graphs , 1986, Discret. Math..

[17]  Joseph W. Goodman,et al.  A generalized convergence theorem for neural networks , 1988, IEEE Trans. Inf. Theory.

[18]  C. Guzelis,et al.  Construction of energy landscape for discrete Hopfield associative memory with guaranteed error correction capability , 2003, First International IEEE EMBS Conference on Neural Engineering, 2003. Conference Proceedings..

[19]  Stella N. Batalama,et al.  An efficient learning algorithm for associative memories , 2000, IEEE Trans. Neural Networks Learn. Syst..