The antiphon: a device for reliable memory from unreliable components. I

A memory device (the antiphon) is proposed which reverberates an excitation pattern (corresponding to a state of an M-bit memory) between two sets of nodes (α- and β-nodes). The β-nodes may behave unreliably (stochastically). It is shown that reduced versions of the antiphon produce two memory mechanisms now classical: the repeated cycling of a message through the steps of encoding, noisy transmission and decoding, and the Hopfield net. However, the antiphon presents several differences of detail from the Hopfield net; it is also more physically explicit and more easily analysed. One mode of operation of the antiphon is analysed here; this analysis leads to an optimization of the statistics of the network and a lower bound for the memory capacity of the antiphon.

[1]  W. A. Little,et al.  Analytic study of the memory storage capacity of a neural network , 1978 .

[2]  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.

[3]  Gérard Weisbuch,et al.  Scaling laws for the attractors of Hopfield networks , 1985 .

[4]  Sompolinsky,et al.  Storing infinite numbers of patterns in a spin-glass model of neural networks. , 1985, Physical review letters.

[5]  Sompolinsky,et al.  Spin-glass models of neural networks. , 1985, Physical review. A, General physics.

[6]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[7]  Geoffrey E. Hinton,et al.  Learning and relearning in Boltzmann machines , 1986 .

[8]  B. Derrida,et al.  Evolution of overlaps between configurations in random Boolean networks , 1986 .

[9]  Peter Whittle,et al.  Systems in stochastic equilibrium , 1986 .

[10]  E. Gardner.  Structure of metastable states in the Hopfield model , 1986 .

[11]  B. Derrida,et al.  Random networks of automata: a simple annealed approximation , 1986 .

[12]  H. Hilhorst,et al.  On the approach of the stationary state in Kauffman's random Boolean network , 1987 .

[13]  J. Nadal,et al.  Learning and forgetting on asymmetric, diluted neural networks , 1987 .

[14]  D. J. Wallace,et al.  Dynamics and statistical mechanics of the Hopfield model , 1987 .

[15]  Santosh S. Venkatesh,et al.  The capacity of the Hopfield associative memory , 1987, IEEE Trans. Inf. Theory.

[16]  Richard E. Blahut,et al.  Principles and practice of information theory , 1987 .

[17]  E. Gardner,et al.  An Exactly Solvable Asymmetric Neural Network Model , 1987 .

[18]  Richard A. Andersen,et al.  A back-propagation programmed network that simulates response properties of a subset of posterior parietal neurons , 1988, Nature.