+Weijian Meng googleblog.blogspot.co.uk/2016/01/alphago-machine-learning-game-go.html We trained the neural networks on 30 million moves from games played by human experts, until it could predict the human move 57 percent of the time 预测走子有57%准确率和职业棋手一样 我又查了一下大概训练了15万局职业棋局,但是自我博弈下的是千万局以上,比重大的多。而且按照他的水平下每盘都是职业局水平。所以经常有看不懂的棋。 说是在模仿人类下棋就太不了解了。
Their supervised learning policy network was trained on 29.4 million positions from 160,000 KGS games between 6-9d strong amateurs. 初始训练数据:16万盘KGS比赛 业余6-9段。 In a similar matchup, AlphaGo running on multiple computers won all 500 games played against other Go programs, and 77% of games played against AlphaGo running on a single computer. 分布式版本对单机版胜率77%,这次和上次用的都是单机版。