Quoting Thomas Lavergne <thomas.laver...@reveurs.org>:
- the best play is a good only if played immediatly and very bad if
played later in the game :
- the first playout for this play resulted in a lost.
score and RAVE score will be very low and this play will never be
considered again until a very long time.
You raise an interesting concern.
The simple solution to your question is to add an exploration term
using UCT for example. Then it becomes an empirical question what
parameter for exploration gives the strongest play. My experience is
that the best parameter is so small it can be set to zero.
I think the conditions you defined are very rarely completely
fulfilled. What can be true often however is that a single bad move
makes the best move very bad if played later in the game. If the bad
move happen to be the second best move, it will be searched a lot
lowering the AMAF score (rw/rc) for the best move.
This is likely to happen when there are several local moves that more
or less solves the same problem. That is when one move is played the
effect of the other move played later will overlap with the first.
-Magnus
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
