June 26, 2012 at 1:08 PM ET
An artificial brain that Googlers built with 16,000 computer processors and a nifty algorithm has taught itself to recognize images of cats posted on the Internet.
While cats provide web surfers endless hours of giggles, the feat of the so-called neural network is a serious step forward in the realm of machine learning, the New York Times reports.
The network is among the largest of its type ever built, thanks in part to falling costs and abundance of computers available for research.
The Googlers fed their artificial brain 10 million images that were randomly selected from YouTube videos. Given a chance to see a random slice of what we put on the Internet, the machine followed our lead and learned to find cats.
“We never told it during the training, ‘This is a cat,’” Google fellow Jeff Dean told the New York Times. “It basically invented the concept of cat.”
The results show that “it is possible to train a face detector without having to label images as containing a face or not,” the team notes in a paper on the research posted on arXiv.org and presented this week at a conference in Scotland.
As we build ever bigger and more sophisticated artificial brains, they could help improve image search, speech recognition and machine language translation – services that Google provides.
John Roach is a contributing writer for msnbc.com. To learn more about him, check out his website and follow him on Twitter. For more of our Future of Technology series, watch the featured video below.