Re: [gnugo-devel] End game strategy

[Arend Bayer]

On Mon, 1 Sep 2003, SP LEE wrote:

> Currently, gnugo enters engame mode when there is no move found of value
> greater than 6. However, there are many typical big end game moves worth 10
> to 30 points. I would think the gnugo should enter end game mode when :
> 1. All b/w groups are non-critical
> 2. There is no more invasion chance to either color
> 3. The only task for both color is to expand or defend territory.
>
> The following situation from the game with go ++ maybe a illustration
>    A B C D E F G H J K L M N O P Q R S T
> 19 . . . . . . . . . . . . . . . . . . . 19
> 18 . . . . O X X . . . . . . . . . O . . 18
> 17 . . . . O O X . . . X . O . O . O X . 17
> 16 . . O + . . X . . + . . . . . O X . . 16
> 15 . . . . . . . . . . . . . . . . X . . 15
> 14 . . . . . . . . . O . . . . . . . . . 14
> 13 . O . . O . O . . . . . . . . . . . . 13
> 12 . X O . . . . . . O . . . . . X . . . 12
> 11 . X O O X . . . . . . . . . . . . . . 11  WHITE (O) has captured 0 stones
> 10 . O X O . X . . X + . . . . . + . . . 10  BLACK (X) has captured 0 stones
>  9 . X X . . . . . . . . . . . . . X . . 9
>  8 . . . . . . . . . . . . . . . . . . . 8
>  7 . . X . . . X . . . . . . . . . . . . 7
>  6 . . . . . O . . . . . . . . O . X . . 6
>  5 . . . . . . . . . . . . . . . . . . . 5
>  4 . . . + . X . X . X . . . . . O . O . 4
>  3 . . . X . . . . . . . . . O . . . . . 3
>  2 . . . . . . . . . O . . . . . . . . . 2
>  1 . . . . . . . . . . . . . . . . . . . 1
>    A B C D E F G H J K L M N O P Q R S T
>
> I think in this case, gnugo should turn off all minimum accepted value such
> as a J pattern and calculate territory value solidly. Gnugo played S18 which
> doesn't worth much, just due to the minimum accepted value of 13.24.

I think the purely territorial value of S18 is around 13 pts, plus it
probably makes M18 absolutely sente, so S18 is objectively not
overvalued.

But I agree that E9 is better (D9 would be even better). The territorial
evaluation is bad here.
The way I think of this problem (it's quite typical) is that the white
move at D9 has so a lot of followups. Remember that the territorial
evaluation is trying to do the following:

- Compare the position as it is, with White moving first in all local
situations, with the position after Black E9, again with White moving
first in all local situations.

However, it cannot really assess the damage of White moving first around
E9. (It probably thinks both the intrusions towards F7 and towars G9 can
be stopped.

I have no ideas how to solve this without major change. It could
probably be solved either with a CGT-style move valuation, or with
global alpha-beta reading.

Arend