[Computer-go] CGT endgame solver
weiqiprogramming at gmail.com
Mon Jul 13 13:37:44 PDT 2015
Even strong engines like Fuego will give wrong sequences in yose.
1. Sometimes it will play a 1 point sente too early (right as it's
available, so maybe it already played it early game), get into a ko and
lose 10 points in seki or die
This is because playouts always give a better result for the engine
playing the 1 point sente than not playing it
2. It doesn't read out ko variations correctly anyway. Sometimes it
makes a thousand year ko instead of a direct ko in a simple corner pattern.
3. It plain doesn't play the correct sequence of moves in the endgame.
In fact, none of the engines really ever agree on a move order in the
endgame and if you run a few simulations at the end of the game you'll
keep getting different results, sometimes W+5.5 and sometimes B+3.5
On 2015-07-13 11:47, Darren Cook wrote:
>> performance at endgames is worse than middle, because IMHO MC
>> simulations don't evaluate the values (due to execution speed) of
>> yose-moves and play such moves in random orders. Assuming there are 7,
>> 3, 1 pts moves left at a end position, for example.
> (Sorry for two messages). I just thought, don't pattern libraries
> capture this? (I may be wrong on this, but I thought all the strong
> programs were using patterns to influence search.) The yose patterns
> come up multiple times in every game, so shouldn't stopping a monkey
> jump get searched before a hane-connect move on the 1st line?
> Computer-go mailing list
> Computer-go at computer-go.org
More information about the Computer-go