Google creates 'computer brain' that loves cats!

The efforts of a team of scientists, hired by the famous Google X Labs, have been successful in the direction of crating the first artificial brain, capable of learning from its own experience.Rather famous for inventing the self-driving cars and for the Google Project Glass (augmented reality glasses), the scientists of Google X Labs are currently working on an new, highly-ambitious project, that attempts to create a virtual replica of the human brain, capable of learning.It turns out our future robot overlords love watching cats on the Internet, too. Google used 16,000 computer processors to create an artificial brain, connected it to the Internet, and fed it random images from 10 million YouTube videos so it could "learn."
The experiment worked thusly: the computers were force-fed thumbnail images culled from 10 million YouTube videos, all of which were selected at random.“The Google brain assembled a dreamlike digital image of a cat by employing a hierarchy of memory locations to successively cull out general features after being exposed to millions of images. The scientists said, however, that it appeared they had developed a cybernetic cousin to what takes place in the brain’s visual cortex.”

The researchers attribute this to part of the brain acting as a “cybernetic cousin” to the visual cortex, constructing an image of cat subconsciously via “grandmother neurons” that are activated when exposed to a particular face over and over again.Overall, Google’s team seems to think they can map out the entire visual cortex in less than a decade. Let’s hope its sophistication isn’t misplaced.

This Google creation performed like an image analyser. It learnt by itself as no one fed it with a description of a cat. Team leader Jeff tweeted, "I've been working on training systems for very large neural networks recently. One cool result we've found is that a large network trained with totally unlabeled data can automatically discover high-level concepts like human faces, cats, etc. (cats because we trained on still images from a large collection of YouTube videos).