Google has performed the difficult task - the invention of an ‘artificial brain’ with the help of 16,000 computer processors, having around a billion links.
The group headed by Google’s Jeff Dean then given it random pictures extracted from 10 million YouTube clips - and let it ‘learn’ by itself.
Incredibly, the device targeted in on cats. “We never ever instructed it throughout the training ‘this is a cat’,” said Dean. “It mainly designed the concept of a cat.”
“Contrary to what seems to be a widely-held intuition, our fresh results uncover that it is feasible to teach a face detector without having to label images as containing a face or not,” states the team.
“We additionally discover that the exact same network is sensitive to other high-level ideas such as cat faces and human bodies.”
“Starting with these learned features, we trained our network to obtain 15.8 percent accuracy in recognising 20,000 object categories from ImageNet, a leap of 70 percent relative improvement over the previous state-of-the-art,” it said, Daily Mail reports.
