Abstract
Numerous studies have been done with neural models of associative memory. Hopfield model is one of autoassociative memory and is deeply studied with the theoretical analysis and numerical simulation by introducing the energy functions. Further it is well known that higher order Hopfield model has higher ability than the conventional one. Bidirectional associative memory (BAM) and multidirectional associative memory (MAM) are heteroassociative models and are also deeply studied using the same method as Hopfield model. However, there are few higher order neural models of heteroassociative memory. Specifically, little is known about higher order heteroassociative memory with the decreasing energy function. This paper proposes higher order MAM (HOMAM) with the energy function. From numerical simulation and static analysis, HOMAM is superior to other model. Specifically, we have shown that the memory capacity of HOMAM is about 0.18P/n 2 and HOMAM has the high ability of error correcting, where P is the number of memorized patterns and n is the number of neurons.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
D. Rumelhart and J. McClelland, Parallel Distributed Processing: Explorations in the Microstructure of Cognition; Vol. 1: Foundations: Vol. 2: Psychological and Biological Models, MIT Press, Cambridge, Mass., 1986.
J. Hertz, A. Krogh and R. Palmer, Introduction to the Theory of Neural Computation, Addison-Wesley Publishing Company, Inc., Redwood City, CA, 1991.
C. Lin and C. Lee, Neural Fuzzy System, Prentice Hall PTR.., 1996.
M. Okada, “Notions of associative memory and sparse coding”, Neural Networks, Vol. 9, pp. 1429–1458, 1996.
S. Amari and K. Maginu, “Statistical neurodynamics of associative memory”, Neural Networks, Vol. 1, pp. 63–73, 1988.
S. Chan and J. Michael, “Memory capacity of artificial neural networks with high order node connections”, Proceedings of ICNN, pp. 207–216, 1988.
D. Psaltis, C. Park and J. Hong, “Higher order associative memories”, Neural Networks, Vol. 1, pp. 149–163, 1988.
S. Yatsuki and H. Miyajima, “Associative Ability of Higher Order Neural Networks”, in Proc. ICNN’97, Vol.2, pp. 1299–1304, 1997.
S. Yatsuki, and H. Miyajima, “Statistical Dynamics of Associative Memory for Higher Order Neural Networks”, IEEE Proc. of ISCAS 2000, Vol.3, pp. 670–673, 2000.
L.F. Abbott, and Y. Arian, “Storage Capacity of Generalized Networks”, Physical Review A, Vol. 36, No. 10. pp. 5091–5094, 1987.
J.J. Hopfield, “Neural networks and physical systems with emergent collective computational abilities”, Proceedings of the National Academy of Sciences, Vol. 79, pp. 2554–2558, 1982.
E. Goles, F. Fogelman, and D. Pellegrin, “Decreasing Energy Functions as a Tool for studying threshold networks”, Disc. Appl. Math, Vol. 12, pp. 261–277, 1985.
S. Yoshizawa, M. Morita, and S Amari, “Capacity of Associative Memory using a Nonmonotonic Neuron Model”, Neural Networks, Vol. 6, pp. 167–176, 1993.
B. Kosko, `Bidirectional associative memory“, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 18, pp. 49–60, 1988.
M. Hattori, M. Hasegawa, “Multimodule associative memory for many-to-many associations”, Neurocomputing, Vol. 19, pp. 99–119, 1998.
P.K. Simpson, “Higher-ordered and intraconnected bidirectional associative memories”, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 20, pp. 637–653, 1990.
C.S. Leung, L.W. Chan and E.M.K. Lai, “Stability and statistical properties of second-order bidirectional memories”, IEEE Transactions on Neural Networks, Vol. 8, pp. 267–277, 1997.
Y. Uesaka and K. Ozeki, “Some properties of associative type memories”, IEICE of Japan, Vol.55-D, pp. 323–330, 1972.
Y. Hamakawa, H. Miyajima, N. Shigei and T. Tsuruta, “On Some Properties of Higher Order Correlation Associative Memory of Sequential Patterns”, submitted to R,ISP.
R. Meir and E. Domany, “Exact Solution of a Layered Neural Network Memory”, Physical Rev. Letter, 59, pp. 359–362, 1987.
S. Amari, “Statistical Neurodynamics of Various Versions of Correlation Associative Memory”, Proceedings of IEEE Conference on Neural Networks, Vol. I, pp. 633–640, 1988.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Miyajima, H., Shigei, N., Kiriki, N. (2004). Higher Order Multidirectional Associative Memory with Decreasing Energy Function. In: Rajapakse, J.C., Wang, L. (eds) Neural Information Processing: Research and Development. Studies in Fuzziness and Soft Computing, vol 152. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-39935-3_8
Download citation
DOI: https://doi.org/10.1007/978-3-540-39935-3_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-53564-2
Online ISBN: 978-3-540-39935-3
eBook Packages: Springer Book Archive