Oh yes, you read that title correct “IBM to roll out all new program SyNAPSE chips inspired by human brains”. Just imagine your situation what if you can get any software which is actually live before you which works under the commands of your neurons of the brain. Oh yes, you might be excited to hear about the same and might be right now fantasizing about various situations, right?? We are living in such an era where the modifications are not only taking place based on the silicon chips but far more and beyond that, the biological entities are being taken as a revolutionary tool to create an envisioned and fascination of millions turned to reality.
Then let me begin with my today’s article about news which I would love to share with you all. Around 2 years back IBM (International Business Machines) announced of a technology which it had developed which was a “cognitive computer chip” motivated with the neural architecture of a human. This project was with the intention of developing a computing system which will be able to handle tasks that will be easy for any human brain to handle but not for computers. It was a included under the SyNAPSE project. Today the same company announced before the media that they are now in the chase for developing a programming framework and architecture where programmers can program these chips and thus can be transformed into a reality and made live and handed into the hands of millions.
The point of developing a whole new base was because the concept of “cognitive computers” is totally new and so it will require a new language which will be far more different than the current programming languages.
It is believed that IBM’s concept of hardwares for “cognitive computers” is built around small “neurosynaptic cores”. These cores are the ones which are modeled on a human brain and it features the following things into it which are “neurons” as the processors, 256 “axons” which is the memory and 64,000 “synapses” which is the nothing else but the communication link between the neurons and the axons. IBM envisions to extend the scale of synapses upto 100 trillion for building a cognitive computer. Sounds like a sci-fi Hollywood film, right?? 😛
Yet these computers are still in the form of an imagination only as its existence relies on the future, but a simulated version of such computer is available which exists on IBM’s “Blue Gene” supercomputers, at Lawrence Livermore and Lawrence Berkeley National Laboratories. With the help of that simulated model the team is developing a new dimension of programming language’s architecture which can eventually be used into the real cognitive computers.
About the language:
It is asserted that this new language is built around the concept called “corelets” which is an object oriented abstraction of each neurosynaptic core. Each corelet has got only 256 inputs as well as same number of outputs. These inputs and outputs are used like the connecting medium between the cores.
Dr. Modha said clarifying the idea more that “Traditional architecture is very sequential in nature, from memory to processor and back. Our architecture is like a bunch of LEGO blocks with different features. Each corelet has a different function, and then you compose them together.”
An instance of this sort of device:
Say if you want to locate a face in a crowd, one corelet might be looking for colours, another one for the shape of the nose, third one identifying the cheekbones, fourth one might be looking at the eye colour and so on. Actually these all corelets timing function will be slow in rate but when the processing will be there in parallel the resulting output will be faster. From this we can say that a cognitive computer in real sense can be used for applications involving pattern recognition and other applications which require shifting through a huge data efficiently from the database which is the current drawback of our traditional computers. Looking at the other end point another truth also reveals that – the cognitive computers might not work out efficiently in other criterias as our traditional computers do currently. This is one of the reasons why Mr. Modha envisions future computers as the hybridization of SyNAPSE chips and traditional computers.
“Today’s computers are very good at analytics and number crunching; think of today’s computers as left brained and SyNAPSE as right brained.” Dr. Modha appended.
If we are think that such computers will be a replaceable technology with a human brain then that is not actually a valid point because after all it’s a machine and not a natural brain.
Dr. Modha said clarifying the things out that “We can’t build a brain, but the world is being populated every day with data. What we want to do is to make sense of that data and extract value from it, while staying true to what can be build on silicon. We believe that we’ve found the best architecture to do that in terms of power, speed and volume to get as close as we can to the brain while remaining feasible.”
The IBM team is expectantly trying to build applications like:
- Using the chips to produce wearable computing glasses which in turn can process visual data to those with impaired vision.
- The chips might be expected to work as search and rescue robots to find injured people in a state of emergencies.
But as Dr. Modha believes and I quote that “The possibilities are endless. The march of time will tell. We’re creating the platform for a huge community that we hope will get involved. Progress never stops – once you’ve scaled one peak, the next one manifests.”
You can see a video explaining the project below:
We just need to wait till then but as far as I think I am eagerly waiting for the first step towards a revolution take its beginning.
So, what do you say?? 🙂
Kindly leave your vision for the same in the form of comments. 🙂
Kindly depart your imaginative and prescient for the same in the form of feedback. 🙂