![8bitfuture:
Google’s neural network teaches itself to identify cats.
Using large-scale neural networking software distributed across 16,000 processor cores in Google data centres, the self-teaching software was able to identify cats after one week of watching YouTube videos.
“Our hypothesis was that it [the neural network] would learn to recognize common objects in those videos. Indeed, to our amusement, one of our artificial neurons learned to respond strongly to pictures of cats. Remember that this network had never been told what a cat was, nor was it given even a single image labeled as a cat. Instead, it discovered what a cat looked like by itself from only unlabeled YouTube stills,”
Google will now try increasing the size of the network to see performance increases exponentially with network size. While the current network has one billion connections, the human brain has around 100 trillion.
Google is building SkyNet](http://25.media.tumblr.com/tumblr_m6dbeb7jgF1qft3eko1_400.jpg)
Google’s neural network teaches itself to identify cats.
Using large-scale neural networking software distributed across 16,000 processor cores in Google data centres, the self-teaching software was able to identify cats after one week of watching YouTube videos.
“Our hypothesis was that it [the neural network] would learn to recognize common objects in those videos. Indeed, to our amusement, one of our artificial neurons learned to respond strongly to pictures of cats. Remember that this network had never been told what a cat was, nor was it given even a single image labeled as a cat. Instead, it discovered what a cat looked like by itself from only unlabeled YouTube stills,”
Google will now try increasing the size of the network to see performance increases exponentially with network size. While the current network has one billion connections, the human brain has around 100 trillion.
Google is building SkyNet
(via knightworg)
