Re: [computer-go] The "standard techniques" of go playing programs

[David Fotland <fotland@xxxxxxxxxxxxxxx>]

This is generally true.  Most of the pattern knowledge is hand entered by a 
human.  This pattern knowlege
is used to select moves for search, since the searches must be highly 
pruned.  There have been attempts to
use machine learning.  Take a look at Neurogo, one of the most 
successful.  I think most programs have
some form of machine learning.  Go++ can learn patterns for move selection 
from professional games, but it
got much stronger when human-selected moves were added.  Many Faces uses a 
database of profession games
to select moves in the fuseki.

You should note that strong chess programs also have their knowledge 
encoded by hand by the programmer.
Machine learning chess programs have never been very strong.

I think a better methodology would be to get copies of programs from 5, 10, 
15 years ago, and test their relative
strength and extrapolate.

I have an archive of old strong go programs, and I could make some 
available to you if you could get permission
from the authors.

Regards,

David Fotland

At 07:02 PM 2/1/2004 +0100, mpe501@xxxxxxxxxxxxxxxxx wrote:
> According to the computer go ladder the four best programs (on 19x19) are: 
> Go4++, GoAhead, ManyFaces and GnuGo.
>
> Would it be accurate to describe these programs as something akin to 
> expert systems, that is, the go knowledge of individual people encoded in 
> program form? Is it correct that they don't use any fancy machine learning 
> techniques but are hand-tuned  by humans(the pattterns etc.) for optimal 
> performance? Each new pattern has to added by a human?
>
> I seem to remember reading something about that at least manyfaces and 
> gnugo work like that....
>
> I am doing an investigation were I try to answer the question "Given 
> incremental improvement on the current standard techniques in go 
> programming, and incremental improvment in computer performance, is it 
> likely that we will ever get a professional level program?".  My working 
> hypothesis is no... I will attempt to test it by taking old versions of 
> e.g. gnugo and see how large the difference in strength is between 
> different vesions when run on the same hardware and also how much better 
> it gets with increased computing power. I will then try to estimate how 
> much better programs are likely to become in the next 5, 10, 20, 50 years.
> However, first I need to know if my assumption that there is such a thing 
> as a set of "standard techniques" in go programming is correct and that 
> they are something similar to what I described above. Are they?
>
> Regards,
> David
>
> _______________________________________________
> computer-go mailing list
> computer-go@xxxxxxxxxxxxxxxxx
> http://computer-go.org/mailman/listinfo/computer-go

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://computer-go.org/mailman/listinfo/computer-go