Researchers at IBM and Cornell University have developed a new computer chip that uses minimal power to perform real-time complex calculations. Dubbed TrueNorth, the chip is based on the design of the human brain.
“Inspired by the brain’s structure, we have developed an efficient, scalable, and flexible non–von Neumann architecture that leverages contemporary silicon technology,” the IBM and Cornell researchers wrote in an article published last week in the journal Science. The chip, which is the size of a postage stamp, is made up of 5.4 billion transistors that are wired to emulate a brain with 1 million neurons that talk to one another via 256 million synapses.
“In 2011, we had a chip with one core”, said Dr. Dharmendra S. Modha, Manager, Cognitive Computing, IBM Research. “We have now scaled that to 4096 cores".
The chip is not aimed to compete with today’s conventional microprocessors, but is rather intended to supplement them. According to Modha, while current von Neumann machines are good at crunching big numbers and doing heavy computational lifting, they aren't good at solving problems in sensing and moving, and this is where these cognitive computers come in, as they are good at processing images, sound, and other sensory data.
The processor may be able to recognize that a woman in a video is picking up a purse, or control a robot that is reaching into a pocket and pulling out a quarter, The NewYork Times notes. Despite the kind of work the chip is designed to do, it draws only a tiny amount of electricity (about 70 mW).
The chip is the product of the Systems of Neuromorphic Adaptive Plastic Scalable Electronics (SyNAPSE) program, which has received $53 million in funding since 2008 from the Pentagon’s Defense Advanced Research Projects Agency (DARPA).