Abstract
Based on one-step Hebbian learning we incorporate higher order correlations into the flexible architecture of a neural network optimized in first order. Sparse connectivity as well as a suitable summation technique eliminate the proliferation problem of higher order terms.
Preview
Unable to display preview. Download preview PDF.
References
W.W. McCulloch and W. Pitts,1943, Bull.Math.Biophys. 5, 115–133
E. Caianiello, 1961, J.theor.Biol. 1, 204–235
D. Psalits, C.H. Parc and J. Hong, 1988, Neural Networks 1, 149–163
C.L. Giles and T. Maxwell, 1987, Applied Optics 26, 23, 2972–4978.
K.E. Kürten, to be published
K.E. Kürten, 1990, J. Phys. France 51, 1585–1594
K.E.Kürten,1990, Parallel Processing in Neural Systems and Computers,World Scientific, ed. by R.Eckmiller, G.Hartmann and G. Hauske, 191–194
M. Bouten, A. Komeda and R. Serneels, 1990, J.Phys.A 23, 2605–2612
B. Derrida, E. Gardner and A.Zippelius, 1987, Europhys.Lett. 4, 167–173.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1990 Springer-Verlag
About this paper
Cite this paper
Kürten, K.E. (1990). Higher order memories in optimally structured neural networks. In: Garrido, L. (eds) Statistical Mechanics of Neural Networks. Lecture Notes in Physics, vol 368. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3540532676_69
Download citation
DOI: https://doi.org/10.1007/3540532676_69
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-53267-5
Online ISBN: 978-3-540-46808-0
eBook Packages: Springer Book Archive