There must be no bias in the selection of the moves (unless of course
this is purposeful and you know what you are doing), they must be
uniformly random - no particular move being more likely than another
to be chosen.
Do something simple to start with to make sure it is working
correctly, then optimize it later. For instance, in the
random games:
for each move you play in the random games:
1. Choose a move randomly from the whole list.
2. if the move is legal play it.
3. If the move is illegal - put it away so it doesn't
get selected again and repeat at step 1.
You can "put a move away", for instance by placing the last
move in the list in it's place and decrement the size of
the list.
- Don
On Mon, 2007-02-05 at 17:33 -0500, Chris Fant wrote:
> On 2/5/07, Don Dailey <[EMAIL PROTECTED]> wrote:
> > Did you look at the games? Sometimes there is something obvious.
> > For instance does it pass too early or lose games on forfeit?
> >
> > Although I can't see your code, I would suspect the following
> > problems:
> >
> > 1. Is the eye avoid routine 100% correct?
> >
> > 2. Are the moves actually uniformly random? You cannot just
> > shuffle the list of moves for instance (without doing other
> > things too.)
> >
> > 3. Make sure you are scoring the end of game correctly.
> >
> > 4. See if black is winning a lot more than white or visa versa.
> >
> > For 3, you must consider komi correctly
> > and you must keep stats on the wins and losses, not on the
> > amount of territory for each side - that would weaken you
> > signficiantly.
> >
> > Here is how my program scores:
> >
> > if ( (bs + bt) - (ws - wt) > komi ) then black_wins else white_wins.
> >
> > bs = black stones
> > bt = black territory
> > ws = white stones
> > wt = white territory
> > komi = integer komi (but it works with fractions too.)
> >
> > - Don
>
> I think so, but I guess I'll have to go over each of these one by one.
> Can you elaborate on #2?
> _______________________________________________
> computer-go mailing list
> [email protected]
> http://www.computer-go.org/mailman/listinfo/computer-go/
_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/