I'm not sure if it's wise to ignore games lost on time. For a MCTS
program it makes sense to adjust the time taken for the move based on
its perceived chance of winning. But that means a program is more
likely to lose on time because it's losing anyway, and that judgement
involves the komi.
Maybe better to ignore games lost on time before a certain number of
moves were played, which may indicate a crash.
Mark
On 8-okt-08, at 23:22, Don Dailey wrote:
I also won't consider games lost on Time for this purpose and I
will use
the ELO calculation from Bayeselo, not the current ratings when the
games were played
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
