Re: computer-go: Programming the capturing game

[Tristan Cazenave <cazenave@xxxxxxxxxxxxxxxxx>]

Bruno Bouzy a écrit :
> 
> Tristan Cazenave wrote:
> 
> > You wrote your solution is depth 15, and the shortest one
> > I found was depth 17... I am quite interested in seeing
> > your solution...
> 
> I carefully read your AGPS paper which is very interesting.
> AGPS algorithm seems to be a too selective search to me because, if my
> reading is correct,
> the diagonal move on the 6x6 cross-cut position leads to a13 ply
> capture sequence.
> :-|

The shortest win I was able to find "by hand" with the diagonal move
is 15 plies. The sgf file is attached. But you are right, it is less 
than the 17 moves found by AGPS. AGPS does not find the shortest win,
but it finds a win faster... Maybe 6x6 is too small to really show
the interest of AGPS, I will give a try to 8x8...

Does someone has a winning sequence in mind for 8x8 with a cross cut
in the centre ?

Tristan.
(;GM[1]FF[3]
RU[Chinese]SZ[6]HA[0]KM[5.5]
PW[White]
PB[Black]
GN[White (W) vs. Black (B)]
DT[2002-02-06]
SY[Cgoban 1.9.11]TM[30:00(5x1:00)];AW[dc][cd]AB[cc][dd];B[bb];W[db];
B[bd];W[ce];B[de];W[be];B[ae];W[cb];B[bc];W[bf];B[df];W[ed];B[cf];
W[ee];B[af]
)