The nice people at Google have put together a “neural network” of computers which is capable of learning.  In three days it learned to spot cats in pictures, even though it had never been told what one looked like.

Ah, neural networks.  Everyone’s favourite method of creating supercomputers and robots.  Say, Matt, how do you feel about doing a post on different types of robot-brain?  Is there enough variety out there to make it interesting?

The thing that really gets me about it, though, is that right at the very end there is this sentence:

As well as spotting cats, the computer system also learned how to pick out the shape of the human body and to recognise human faces.

A one line throw-away at the end of the article?  I guess they thought people would be more impressed by kitties

Advertisements