A human plays a computer in a go 囲碁 match. After each game the komi is adjusted depending on the outcome. The komi should be the greatest handicap at which the human has yet to beat the computer.
Different komi for white and black.
This especially works well if the computer is deterministic (as it should be). There is no need to play again if the human has already beaten the computer at a given handicap; the human can simply repeat the moves to win again.
No comments :
Post a Comment