Information
Analog Neural Computer
During his Ph.D. at the University of Pennsylvania (Jan Van der Spiegel and Paul Mueller, PI), Prof. Ralph Etienne-Cummings was one of the designers of the first large scale neural computer implemented using VLSI chips.1 He was responsible for designing the neurons and synaptic time-constants chips, and for system integration. Recently, we have started looking at neural computing again, but this time computation with spikes, instead of rate codes.2
References
- ["A General Purpose Analog Neural Computer and a Silicon Retina for Real Time Target Acquisition, Recognition and Tracking," CAMP 93, M. Bayoumi, L. Davis and K. Valavanis (Eds.), pp. 48-57, 1993; "An Analog Neural Network with Modular Architecture for Real-Time Dynamic Computations," IEEE J. Solid-State Circuits, Vol. 27, pp. 82-92, 1992; "Design and Performance of a Prototype General Purpose Analog Neural Computer," Neurocomputing, Vol. 4, pp. 311-324, 1992]
- ["Real-Time Image Processing using a Spiking Imager and an Integrate-and-Fire Array Transceiver System," accepted to Neural Computation, Fall 2006; "A Real-Time Spike Domain Sensory Information Processing System," IEEE ISCAS 2005, Kobe, Japan, May 2005; "Saliency-Driven Image Acuity Modulation on a Reconfigurable Silicon Array of Spiking Neurons," Proc. of Neural Information Processing Systems, Vancouver, Canada, December 2004]