> From: Petr Baudis <[email protected]>
> On Tue, Sep 07, 2010 at 09:09:51AM -0700, Dave Dyer wrote:
> > I take the observation that human players lose at first, then rapidly
> > learn
>to win against the best programs to be evidence that the programs have
>systematic weaknesses.
>
> I think we have no good, statistically meaningful statistics to
> confirm this observation so far. It would be very nice if someone looked
> at this more seriously and examine this with real scientific rigor.
> There is a large body of games against a lot of random people (and many
> of them being notorious bot-players) available in the KGS archives.
>
> > The mono-culture that is developing, with all programs using variations on
>monte carlo, is just driving the herd (of programs) up a blind alley.
>
> So what would you do about that? ;-)
Lay aside the statistical mindset and review lost games. Once a game reaches a
certain point, there's no statistics involved; it's the other guy's game to
lose
by making a mistake.
The question is, what are those human players learning, to defeat the programs?
_______________________________________________
Computer-go mailing list
[email protected]
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
