Abstract

This paper proposes Neural Network architecture for implementing associative memory. A new model has been developed that has good learning structure and high storage capacity. Hybrid Network Learning comprises interactive counter propagation network and associative memory. Interactive counter propagation network is used for pattern completion. The associative memory is applied for pattern association. Associative memory is content-addressable structure that maps a set of input patterns to a set of output patterns. Associative memory has been expressed in terms of Turing machine. Turing machine is a computing machine which is capable of finding the memory vector which most closely correlates to the input vector. It retrieves previously stored pattern that resembles the current pattern. The Turing machine structure is implemented using B-tree (Turing Tree). The experimental results show that the proposed approach has attained good performance in terms of speed and efficiency.

References

• D. Gnanambigai, P. Dinadayalan and R. Vasanthakumari, ‘‘Associative Learning Network with Turing Machine”, Proc. of the First International Conference on Data Engineering and Management, ICDEM, Tiruchirappalli, 2008, pp. 365-368.
• T. Kohonen, “Self-Organisation And Associative Memory”, Springer-Verlag, 1984.
• John E. Hopcroft, Jeffery D. Ullman, “Introduction to Automata Theory, Languages, and Computation”, NPH, 1994.
• B Kosko, ‘‘Adaptive Bidirectional Associative Memories,’’ Appl. Optics., vol. 26, no. 23, 1987, pp. 4947-4959.
• M. Anabda Rao, “Neural Networks – Algorithms and Applications”, NPH,2003.
• T. Kohonen, “Associative Memories: A System Theoretical Approach”, Springer-Verlag, Berlin, 1977.
• J Austin, ‘‘ A Review of RAM based Neural Networks”, Proc. of the Fourth International Conference on Microelectronics for Neural Networks and Fuzzy Systems., IEEE Computer Society Press, Turin, 1994, pp. 58-66.
• B Widrow, R. G. Winter and R. A Baxter, ‘‘Learning Phenomena in Layered Neural Networks,’’ IEEE 1st International Conference on Neural Networks, M Caudill and C Butler (Ed.), 1987, pp.II-411 - II-429.
• J Austin and T J Stonham, ‘‘An Associative Memory for use in Image Recognition and Occlusion Analysis,’’ Image and Vision Computing, vol. 5, no. 4, Nov. 1987, pp. 251-261.
• HINT81 Hinton, G. E., and Anderson, J. A., (Eds.) “Parallel Models of Associative Memory”, Lawrence Erlbaum Associates, 1981.
• HOPF82 Hopfield, J. J., "Neural networks and physical systems with emergent collective computational abilities", Proceedings of the National Academy of Sciences, Vol. 79, pp. 2,554-2,558, April, 1982.
• Bernard M. Moret, “The theory of Computation”, First edition, PEA, 1998.
• Dmitry O. Gorodnichy. "The Influence of Self-Connection On The Performance of Pseudo-Inverse Autoassociative Networks", "Radio Electronics. Computer Science. Control" journal,, Vol. 2, No. 2, pp. 49-58, 2001.
• Yingquan Wu and Dimitris A. Pados," A Feedforward Bidirectional Associative Memory'', IEEE Transactions on Neural networks, vol. 11, No.4, July 2000.
• Tae-Dok Eom, Changkyu Choi, and Ju-Jang Lee, "Generalized Asymmetrical Bidirectional Memory", Machine Intelligence & Robotic Control, Vol. 1, No. 1, 43–45, 1999.
• H. Shi, Y. Zhao, and X. Zhuang, “A general model for bidirectional associative memories,” IEEE Trans on Systems, Man, and Cybernetics, vol. 28, no. 4, pp. 511–519, 1998.
• Z. Wang, ``A bidirectional associative memory based on optimal linear associative memory,'' IEEE Transactions on Computers, vol. 45, no. 10, pp. 1171--1179, 1996.