Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

He knows it beat the european champion 5-0, and that it outperforms all commercial and free Go games.


Which was meaningless at the time, and I don't think he knew that game beat the European champ until shortly before his game. The difference between the two champions was so large that Lee would be expected to win 95% of the time. All other Go AI was exceptionally weak when compared to the top level players.


It may have been meaningless, but it's all the information the researchers had either. I don't think they misrepresented anything.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: