The world of computing just make a heck of a mass more exciting thanks to IBM ’s implausibly powerfulbrain - inspired chipwhich was unveiled on Thursday . While their prototype single - core system , released back in2011as part of the Systems of Neuromorphic Adaptive Plastic Scalable Electronics ( SyNAPSE ) project , was telling , this newfangled chip drift the quondam one out of the water .
The human brain tops the computing chart as the most efficient organisational arrangement in the globe , so it ’s no wonder IBM and collaborators chose to emulate its capabilities for their new system . This so - called “ cognitive computer science ” aims to mimic the mental capacity ’s ability for perception , action and noesis .
Traditional computing systems can be likened to theleft brain ; they ’re analytic and mathematical with superb number - crunching ability . But if we want something that could be used in more sophisticated arrangement , say artificial intelligence , then we ’re going to need something more right brain - like as well , which is exactly what IBM have been working towards .
This Modern micro chip , which is the sizing of a postage stamp - stamp , addresses the right learning ability social function ofsensory processing and pattern recognition . The thought is to be capable to process , respond to and “ hear ” from information reap from the environs . If successfully meld with a traditional “ left brain ” system , which is what IBM will be attempt over the coming year , we could have a “ holistic computation intelligence ” with Brobdingnagian capabilities in our manpower .
The product , which has been coined “ TrueNorth , ” achieves this through a staggering connection of 1 million programmable neuron , 256 million configurable synapsis ( connection ) and over 4,000 neurosynaptic cores . To put this into linear perspective , the prototype had just 256 neurons , 260,000 synapsis and one core . That ’s a giant bound in just 3 long time . According towired.com , these nerve cell , or “ spiking neurons , ” basically allow the Saratoga chip to encode data as convention of pulses , which is much like one of the many ways scientists believe the learning ability stores info . Details of the chip can be found inScience .
IBM has put the abilities of this chip to the test in variousartificial intelligence tasks , such as simulacrum realization . One test , for good example , involved presenting the chip with a variety of images , and it was incur to be capable to recognize a variety of physical object with around 80 % accuracy . signally , the organization was able to do all of this on just 63 mW of king .
Like the 2011 variant , this check is just a paradigm . IBM hopesto eventually produce a neuro - synaptic cow chip system with a whopping 10 billion neurons and 100 trillion synapsis that can sue entropy whilst consume just 1 kW of power . finally , multiple chips will be strung together on a micro chip board to create a hugenetwork .
IBM envisagesthat the applied science could have a variety of applications , such as sight aid for the blind , health monitoring and transportation such as self - driving cars .
[ ViaAFP , WiredandIBM ]
