This More Powerful Version of AlphaGo Learns On Its Own

See the original posting on Webmonkey

DeepMind unveils a new, more powerful version of AlphaGo that learns without studying human play.