Nabil Imam, Thomas A. Cleland, et al.
Frontiers in Neuroscience
Inspired by the brain's structure, we have developed an efficient, scalable, and flexible non-von Neumann architecture that leverages contemporary silicon technology. To demonstrate, we built a 5.4-billion-transistor chip with 4096 neurosynaptic cores interconnected via an intrachip network that integrates 1 million programmable spiking neurons and 256 million configurable synapses. Chips can be tiled in two dimensions via an interchip communication interface, seamlessly scaling the architecture to a cortexlike sheet of arbitrary size. The architecture is well suited to many applications that use complex neural networks in real time, for example, multiobject detection and classification. With 400-pixel-by-240-pixel video input at 30 frames per second, the chip consumes 63 milliwatts.
Nabil Imam, Thomas A. Cleland, et al.
Frontiers in Neuroscience
Wei-Yu Tsai, Davis R. Barch, et al.
IEEE TC
Andrew S. Cassidy, Paul Merolla, et al.
IJCNN 2013
Alexander Andreopoulos, Brian Taba, et al.
IBM J. Res. Dev