Air Force Research Lab teaming with IBM on advanced neural-network chips

WRIGHT-PATTERSON AIR FORCE BASE, Ohio and ARMONK, N.Y. The U.S. Air Force Research Laboratory (AFRL) has announced that it is collaborating with computing giant IBM on a novel brain-inspired supercomputing system powered by a 64-chip array of the IBM TrueNorth Neurosynaptic System.

researchers believe the brain-inspired, neural network design of TrueNorth will operate more efficiently when handling pattern recognition and integrated sensory processing than systems powered by conventional chips. AFRL is investigating applications of the system in embedded, mobile, and autonomous settings where size, weight, and power (SWaP) is a concern; these uses could include and unmanned aerial vehicles (UAVs).

The advanced pattern recognition and sensory processing power of the scalable platform IBM is building for AFRL will offer the equivalent of 64 million neurons and 16 billion synapses, while the processor component will consume the energy equivalent of a dim light bulb, using only 10 watts to power.

AFRL will combine IBM TrueNorth Neurosynaptic System's real-time -- a "right-brain"-like capability of the system -- with the "left-brain" symbol processing capabilities of conventional computer systems. The large scale of the system will enable both "data parallelism," where multiple data sources can be run in parallel against the same neural network, and "model parallelism," where independent neural networks form an ensemble that can be run in parallel on the same data.

Daniel S. Goddard, director, information directorate, U.S. Air Force Research Lab, says of the project: "AFRL was the earliest adopter of TrueNorth for converting data into decisions. The new neurosynaptic system will be used to enable new computing capabilities important to AFRL's mission to explore, prototype, and demonstrate high-impact, game-changing technologies that enable the Air Force and the nation to maintain its superior technical advantage."

Topics covered in this article