Home News The Artificial Brain built by IBM now has 64 Million Neurons and will have 10 Billion by 2020

The Artificial Brain built by IBM now has 64 Million Neurons and will have 10 Billion by 2020

by Harikrishna Mekala

The system includes 64 of IBM’s TrueNorth chips goes inside a standard server shelf and will be capable of scaling to half a billion artificial neurons with its modern architecture. The TrueNorth chips are distinctive from CPUs in that particular core can run in parallel and without a clock. The operation is essentially flexible, If one core stops running the rest can continue operating without interference.

Just like an organic brain, the system is created for extreme energy efficiency, It needs only 10 watts to power all 64 million neurons and 16 billion synapses. The method isn’t yet as efficient as a biological method but it is getting solid, for illustration, the human brain holds 100 billion neurons and uses about 20 watts.

The ultra-low energy usage will allow vastly more intelligent mobile artificial intelligence operations for use in self-driving cars, smartphones, and aircraft which is the Air Force’s main matter. IBM has done impressive growth over the last six years with its TrueNorth chip architecture, far exceeding the pace of Moore’s Law.

At a lecture last year the head of IBM’s Brain-Inspired Computing division, Dharmendra Modha articulated about his terminal goal of a human brain equivalent in a two-liter box. By 2020 Dr. Modha thinks his team will have made a 10 billion neuron equal system that can fit in a two-liter box and need only 20 watts to operate, that would provide them a desktop operation with roughly 10% the computing ability of the human brain that could simply run on a smartphone battery. Dr. Modha also stated that in order to reach this goal by 2020 he just needs 7nm chips, the FABs for which are now under construction and will be supplying 7nm chips in 2018, two years before of Dr. Modha’s schedule. And IBM has demonstrated that 5nm chips are achievable in the previously.

Take your time to comment on this article.

You may also like