论文信息 - A new neural associative memory with learning

A new neural associative memory with learning

A new learnable associative neural memory has been designed, analyzed, and implemented. The chip implementation is made by using the CMOS current-mode circuits with analog current multiplier and divider. This neural memory can automatically learn and correlative things and has an on-chip ratio-type memory. It can be refreshed by reminding. An experimental three-neuron-net chip was fabricated on a 5-mm*5-mm chip. Experimental results are presented.<<ETX>>

Chung-Yu Wu | Jeng-Feng Lan

[1] Bing J. Sheu,et al. Design of a neural-based A/D converter using modified Hopfield network , 1989 .

[2] S. Grossberg. Nonlinear difference-differential equations in prediction and learning theory. , 1967, Proceedings of the National Academy of Sciences of the United States of America.

[3] Michael Smith. An analog integrated neural network capable of learning the Feigenbaum logistic map , 1990 .

[4] T. Yamada,et al. A self-learning neural network chip with 125 neurons and 10 K self-organization synapses , 1990, Digest of Technical Papers., 1990 Symposium on VLSI Circuits.

[5] J. J. Paulos,et al. Artificial neural networks using MOS analog multipliers , 1990 .

[6] W. Hubbard,et al. A programmable analog neural network chip , 1989 .

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

[8] Teuvo Kohonen,et al. Self-Organization and Associative Memory , 1988 .