[spam probable] [computer-go] Anchor Player

[alain Baeckeroot]

Le mercredi 13 décembre 2006 05:53, Don Dailey a écrit :

> Does a 1 kyu difference mean I can give you 1 stone if I am better and
> expect to come out about even?
yes, 1 handi is 0.5 komi.
> 
> Does this all work out in a transitive way?  If a 6 kyu can give a 7
> kyu 1 stone, and the 7 kyu can give an 8 kyu 1 stone, can the 6 kyu
> expect to play even with the 8 kyu player giving 2 stones?
yes, and it works surprisingly well.

> Would this simple system work:
> 
>    1. Start all players out at the same kyu rating.
> 
>    2. Pair randomly.
> 
>    3. If you win your match, modify kyu rating slightly down.
> 
>    4. If you lose your match, slighly change kyu upward.

Kgs works like this (with more subttle algorithm).
> 
>    
> All this is applied on top of handicaps of course.
> 
> But unless 2 players  are an integer kyu apart, a handicap would be
> slighly
> unfair to one side or the other.  Is it sufficient to modify the
> ratings in linear proportion to the amount of "unfairness?"

Less than 1k difference is nothing for weak players. It is only
meaningful for strong players (several dans or pro)
The link below is stats on even games from European Go Federation
 http://gemma.ujf.cas.cz/~cieply/GO/statev.html

As GNU Go is rated 6k on kgs , this should give more than 30%
for a 9k to beat gnugo in even games.

The traditional way for adjusting handicap needs 3 win in a row (this
is rather difficult)
The fun way is changing handicap after each game (for human
the psychlogical part is very important, one can manage to lose
with many handi due to emotive factor or desire of revenge ...)

Maybe for computer the handicap could be remembered between 2 oppononents,
and the global rank estimated from this ?

GNU Go does not eat memory, even at level 10 it is small and rather fast.
At level 0 it is very poor in reading (rated 2k below level10 gnugo on kgs)
but level 8 should be rather good.
On cgos 9X9 i checked the first 100 000 games of GNU Go 3.7.4 and found
less than 10 nearly nearly identical games (against viking) and less than 5
were rigorously identical. So i bet on 19X19 this will not happen at all.

my 2 cents.
alain