Google: a virtual “brain” to detect... cats

In its mysterious “X Lab” facility, Google is working on a simulation of the human brain. By feeding millions of YouTube thumbnails to the system, the advanced algorithms used in the experiment have been able to identify and detect the “cat” concept all by themselves.

Mountain Views’ “brain simulator” is equipped with 1,000 computers and 16,000 computational units (ie processors) with 1 billion connections overall: the system is used to run advanced machine learning algorithms developed by Google researchers to search and analyze huge amounts of data.

Fed with 10 million images corresponding to as many YouTube videos thumbnails, the algorithms were able to recognize a cat’s face – clearly one of the most pervasive units of digital information clogging the Internet (before, between and likely after YouTube existence) during the last two decades.

“We never told it during the training, ‘This is a cat’’ It basically invented the concept of a cat” said Jeff Dean, one of the project leaders. Google is pretty pleased by the experiment’s results too, it seems, so much that the company decided to relocate the project within its main (and not-so-secret as X Lab) laboratory working on on-line search.

The hope is that machine learning algorithms would become advanced enough to be useful to analyze huge amounts of data – the same type of information overload users go though everyday while searching on-line.

Source: Mashable