“reinforcement learning”

DeepMind’s team built “reinforcement learning” into the programme, meaning the machine played against itself and adjusted its own neural networks based on trial and error. AlphaGo is capable of narrowing down the search space for the next best move from the near-infinite to something more manageable. It can also anticipate long-term results of each move and predict the winner.

http://www.theguardian.com/technology/2016/mar/09/google-deepmind-alphago-ai-defeats-human-lee-sedol-first-game-go-contest

Ume先輩のfollow-upです。１局目で韓国のイ・セドルを下したAlpha Goというソフトは，人工知能の「ディープラーニング（深層学習）」という手法が使われています。その「深層学習」は，こう表現されていました。（Kawada）

http://d.hatena.ne.jp/A30/20160205/1454620851
http://d.hatena.ne.jp/A30/20160308/1457426973
http://d.hatena.ne.jp/A30/20160309/1457523337