Chips for brains


IBM is developing computer chips designed to replicate the way in which human brains operate

IBM researchers in the US have unveiled a new generation of experimental computer chips designed to emulate the brain’s abilities, that should out-perform current computers, and do it all using orders of magnitude less power consumption and space.

IBM’s neurosynaptic computing chips are a complete departure from traditional concepts in designing and building computers. Instead of being based on arrays of transistors, which are either on or off, the chips work in a similar way to the brain’s neurons and synapses. Its first two prototype chips have already been produced and are currently undergoing testing in IBM’s labs. They have already successfully demonstrated simple applications such as navigation, machine vision, pattern recognition, associative memory and classification.

Called cognitive computers, systems built with these chips won’t be programmed in the same way that traditional computers are today. Rather, cognitive computers are expected to learn through experiences, find correlations, create hypotheses, and remember – and learn from – the outcomes, mimicking the brain’s structural and synaptic plasticity.

Dharmendra Modha, project leader for IBM Research said, “These chips are another significant step in the evolution of computers from calculators to learning systems, signalling the beginning of a new generation of computers and their applications in business, science and government.”

The first system has just 256 neurons and 262,144 synapses, but IBM’s long-term goal is to build a system with ten billion neurons and one hundred trillion synapses, while consuming merely one kilowatt of power and occupying less than two litres of volume.

For further information about IBM Research, please visit IBM Research

This article was orginally published on the IBM Vision Community website. IBM are launch partners of LondonlovesBusiness.com.