[computer-go] Explanation to MoGo paper wanted.

[Magnus Persson]

A long time ago ago I spent a few hours on writing a simple chess 
program doing
UCT-search. I got to the point where it actually played better than random but
not very much.

It sort of reminded me of the strength of plain MC in 19x19 Go. The problem is
that many games become very long in chess as for 19x19. My implementation was
also very buggy and some rules of chess were omitted for simplicity. I think
one need to use code similar to what is used for quiscience search to 
make good
heavy playouts. Also using endgame databases (or other techniques) to 
terminate
playouts early (saving time) with a correct result may have a large impact.

I think a good chessprogrammer should be able to improve the strength of the
playouts a lot compared to uniform random moves and a good such chess
UCT-program might be positionally strong at the expence of tactical accuracy.

But those things would be too complicated for me so I stopped working on it.

Quoting Don Dailey <drd at mit.edu>:

> I actually have a working chess program at a fairly primitive stage
> which would be appropriate for testing UCT on chess.
>
> My intuition (which is of course subject to great error) tells me that
> it won't pay off.   However, I'm still quite curious about this and will
> probably give it a try at some point.
>
> To give it a fair chance, you couldn't just quickly whip something
> together,  I think you would have to spend a lot of time experimenting,
> tweaking, debugging, etc to get a good sense of whether it would do the
> job.
>
> Random moves in Chess are problematic.  In a Go position, you can play
> 10,000 random games and this by itself is a rather reasonable evaluation
> function.
>
> Of course the better GO programs impose some bias and control over those
> games and I believe this would be absolutely crucial in chess.  What
> Chrilly calls mutual stupidity doesn't apply as strongly in Chess (in my
> opinion.)    In other words, if you have a somewhat better position,
> sampling a number of random games will not reliably measure this.
>
>
> - Don
>
>
>
>
> On Tue, 2007-07-03 at 05:09 -0700, steve uurtamo wrote:
>> > We felt also, that even if it works, the improvement
>> > measured in Elos would not be very spectacular. The Elo/Effort 
>> ratio is low.
>> > I was simply too lazy (or too professional) to give it a try.
>>
>> it might be fun (even from a non-FPGA point of view) to try it just
>> to see where it lies versus a convential piece of code on equivalent
>> hardware.
>>
>> the game length is roughly the same, or smaller, and the number
>> of move choices is quite a bit more limited than a 19x19 go board,
>> (although larger than a 9x9 board in the sense that in the endgame
>> the board is often fairly empty rather than full) so it might be 
>> surprisingly
>> successful.
>>
>> s.
>>
>>
>>
>>
>>
>>
>> ____________________________________________________________________________________
>> Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's 
>> updated for today's economy) at Yahoo! Games.
>> http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow
>> _______________________________________________
>> computer-go mailing list
>> computer-go at computer-go.org
>> http://www.computer-go.org/mailman/listinfo/computer-go/
>
> _______________________________________________
> computer-go mailing list
> computer-go at computer-go.org
> http://www.computer-go.org/mailman/listinfo/computer-go/
>



-- 
Magnus Persson
Berlin, Germany