Google teaching computers to mimic human brain

8220;Our hypothesis was that it would learn to recognise common objects in those videos,8221; the researchers said.

8220;Indeed,to our amusement,one of our artificial neurons learned to respond strongly to pictures of8230; cats,8221; they continued.

8220;Remember that this network had never been told what a cat was,nor was it given even a single image labeled as a cat.8221;

The computer,essentially,discovered for itself what a cat looked like,according to Dean and Ng.

The computations were spread across an 8220;artificial neural network8221; of 16,000 processors and a billion connections in Google data centers.

The small-scale 8220;newborn brain8221; was shown YouTube images for a week to see what it would learn.

8220;It 8216;discovered8217; what a cat looked like by itself from only unlabeled YouTube stills,8221; the researchers said.

8220;That8217;s what we mean by self-taught learning.8221;

Google researchers are building a larger model and are working on ways to apply the artificial neural network approach to improve technology for speech recognition and natural language modeling,according to Dean and Ng.

8220;Someday this could make the tools you use every day work better,faster,and smarter,8221; they said.

Dean and Ng conceded that there is a long road ahead,since an adult human brain has around 100 trillion connections.

Google X Lab headed by company co-founder Sergey Brin is known for its work on innovations such as a self-driving car and 8220;Terminator8221; film style glasses that provide Internet information about what is being seen.