The Birth of Cognitive Computing Technology

The Birth of Cognitive Computing Technology
Welcome to a Vacuum Cleaner Battery specialist of the iRobot Battery
IBM’s $58 million baby is really the tip of learning computers.
For too long — likely because of inferior programming abilities — scientists have tried to ‘program’ computers to think like us. But this has its limitations.
Take this robot here. Andrew Ng, founder of Coursera, has been teaching this robot to recognise what a coffee cup is through deep learning. But robo didn’t learn to do it on it’s own. It was programmed from labelled data sets. By showing the robot 50,000 pictures of coffee cups and then coding it, the robot was taught by the programmers what a coffee cup looks like.
Personally, I didn’t even know there were 50,000 pictures of coffee with battery like iRobot Roomba 80501 Battery, iRobot Roomba 500 Battery, iRobot Roomba 510 Battery, iRobot Roomba 530 Battery, iRobot Roomba 560 Battery, iRobot Roomba 580 Battery, iRobot Roomba 600 Battery, iRobot Roomba 700 Battery, iRobot Scooba 5800 Battery, iRobot Scooba 5900 Battery, iRobot Scooba 6000 Battery, iRobot Scooba 350 Battery cups to be found on the web.
And this sounds tedious. I mean, did your parents show you 50,000 coffee cups growing up? Or did you just work out what they were by accessing the memory function of your brain?
The real potential is to change from programming behaviours to programming a computer that learns behaviours.
Again, here’s another robot. Or at least the arm of one. In this video here, you can see the arm is catching all sorts flying objects. The robot easily catches a tennis racket.
This robot isn’t a learning robot. It’s shown how to catch a baseball by the operator moving its arm. Because of the coding, it ‘remembers’ the task and then catches the baseball a couple of throws later.
It’s programming by demonstration. Sure it’s a step forward for adaptive robotics. First someone still has to write the code. The robot has learnt how to catch an object. But not because it is a learning robot.
This is why the SyNAPSE chip will have such a huge impact on our technology.
And before you have nightmares of an operating system like V.I.K.I. from the movie I, Robot placing us under house arrest, don’t worry.
The chip does mimic the brain. But only a minute scale.
Really, the chip is it about how the brain shares information.
Another way of looking at it is, it won’t lead to computers thinking for themselves. Rather, based on a set of programmable parameters, computers will be able to act on its own.
As Dharmendra Modha, chief scientist of IBM says ‘We have not built a brain…but we have come the closest to creating learning function and capturing it in silicon in a scalable way to provide new computing capability that was not possible before