Artificial Intelligence Learns to Walk

Google's artificial intelligence entity is actually an algorithm on a graphics computer. It's colloquially called DeepMind after the company that runs it. The operators gave it three different forms and instructions to get from here to there, but didn't tell it how. The algorithm learned to walk all on its own! Sure, it's not at all graceful, but grace was not its goal. Like they say, if it looks stupid and it works, it's not stupid. But it is fun to watch, like a Looney Tunes cartoon or the game QWOP.

(YouTube link)

The humanoid form is the funniest. It had arms but no idea what to do with them, since they don't help it get from here to there. Still, you have to give props to a machine that can learn that well. Next, it will learn all our jobs. -via Metafilter


Login to comment.
Click here to access all of this post's 0 comments
Email This Post to a Friend
"Artificial Intelligence Learns to Walk"

Separate multiple emails with a comma. Limit 5.

 

Success! Your email has been sent!

close window
X

This website uses cookies.

This website uses cookies to improve user experience. By using this website you consent to all cookies in accordance with our Privacy Policy.

I agree
 
Learn More