Fully parallel stochastic computation architecture

A space-efficient fully parallel stochastic computation architecture is described. It circumvents the main drawback of stochastic computation architectures that have been used up to now: the absence of a space-efficient technique of adding weighted input signals in parallel.

[1]  John Shawe-Taylor,et al.  Generating binary sequences for stochastic computing , 1994, IEEE Trans. Inf. Theory.

[2]  Yasuji Sawada,et al.  Functional abilities of a stochastic logic neural network , 1992, IEEE Trans. Neural Networks.

[3]  Alan F. Murray,et al.  Pulse-stream VLSI neural networks mixing analog and digital techniques , 1991, IEEE Trans. Neural Networks.

[4]  Leopoldo García Franquelo,et al.  Fully parallel summation in a new stochastic neural network architecture , 1993, IEEE International Conference on Neural Networks.

[5]  L.G. Franquelo,et al.  Constrained Hopfield neural network for real-time predictive control , 1994, Proceedings of IECON'94 - 20th Annual Conference of IEEE Industrial Electronics.

[6]  Brian R. Gaines,et al.  Stochastic Computing Systems , 1969 .

[7]  Kimmo Kaski,et al.  An effective training method for fully digital pulse-density modulated neural network architecture , 1991, 1991., IEEE International Sympoisum on Circuits and Systems.

[8]  Leopoldo García Franquelo,et al.  VLSI implementation of a fully parallel stochastic neural network , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).