IEEE Journal of Solid-State Circuits · 1996 · 10 citations · 10 references
EngineeringComputer EngineeringComputer ScienceComputer ArchitectureComputer NetworksSystem-on-chip DesignAccelerator ArchitectureData ParallelismField-programmable Gate Array ArchitectureNetwork-on-chip DesignParallel ArchitecturesParallel Computer ArchitectureData Stream ProcessingFloating-point ArithmeticFloating-point InstructionsNeuromorphic HardwareStreaming ArchitecturesStreaming Computation
This paper describes a digital neural network chip for high-speed neural network servers. The chip employs single-instruction multiple-data stream (SIMD) architecture consisting of 12 floating-point processing units, a control unit, and a nonlinear function unit. At a 50 MHz clock frequency, the chip achieves a peak speed performance of 1.2 GFLOPS using 24-bit floating-point representation. Two schemes of expanding the network size enable neural tasks requiring over 1 million synapses to be executed. The average speed performances of typical neural network models are also discussed.
10
Learning representations by back-propagating errors
David E. Rumelhart, Geoffrey E. Hinton, Ronald J. Williams · Nature · 1986 · 30.9K citations
Teuvo Kohonen · Proceedings of the IEEE · 1990 · 8.2K citations