论文信息 - Encoding resolution in Hopfield associative memory networks

Encoding resolution in Hopfield associative memory networks

The number of precision bits for data storage are limited in the hardware implementation of Hopfield type networks. This paper addresses the problem of selection of encoding resolution on the synaptic weights. We investigate the effect of quantization in binary and analog Hopfield networks as a function of data noise and the number of storage patterns. The performance of the networks in terms of recall efficiency was estimated by simulations.

I. Voitovetsky | H. Guterman

[1] Hiroshi Ueda,et al. Associative Neural Network Models Based on a Measure of Manhattan Length (Special Section on the 5th Karuizawa Workshop on Circuits and Systems) , 1993 .

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

[3] John J. Hopfield,et al. Simple 'neural' optimization networks: An A/D converter, signal decision circuit, and a linear programming circuit , 1986 .

[4] J J Hopfield,et al. Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[5] Masahiko Morita,et al. Capacity of associative memory using a nonmonotonic neuron model , 1993, Neural Networks.

[6] Kanter,et al. Associative recall of memory without errors. , 1987, Physical review. A, General physics.

[7] J. Hopfield,et al. Computing with neural circuits: a model. , 1986, Science.

[8] 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.