Bayesian Optimization in AlphaGo improves self-play win-rate from 50% to 66.5%(arxiv.org)3 points by teekno 7 years ago | 0 commentsNo comments yet