Re: computer-go: Learning from existing games

[frank-steinmann@xxxxxxxxxxx (Frank Steinmann)]

> If there were a simple way to tell a good move from a bad move, you
> would just implement it in your program, you wouldn't need the training
> strategy.

Well, knowing if the move is good when you play it and knowing if the move
was good when you have finished the game, are two different things. There
could be a way to use not only the result but the development of the game
from the specific move till the final position as information.

Frank