Re: [Bug-gnubg] Fwd: Neural Net program for For Traditional Iranian/Turkish Backgammon - from Rick Janowski

[Nikos Papahristou]

I have trained neural networks for the Greek backgammon variants plakoto 
and fevga. It seems like neural networks can capture much of the 
complexity of these variants even with few expert features.
You can download a program that plays these variants from my webpage: 
http://csse.uom.gr/~nikpapa/software.html
For now, there is no help, but if you are familiar with gnubg and the 
rules of the games, you shouldn't have any problems :)

You can also see the paper describing the training process in the 
publications section.

Cheers,
Nikos Papahristou

On 28/6/2011 23:17, Joseph Heled wrote:
> A far more interesting and difficult task would be to go for one of
> the other variants In my family they used to play machboosa
> (surrounded), which is the same as the Greek Plakoto, which I find
> strategically orders of magnitude harder. I can hardly tell if I make
> sensible moves or not, as this is a game where you have to slow
> yourself down usually not just a race to the line that is backgammon.
> I even have doubts that a neural net can capture this complexity,
> since most consequences are far ahead in the future - it is like
> playing a backgame all the time, and we know how poor computer players
> are at that, GNUBG especially.
>
>   -Joseph