[Computer-go] Are 4 'easy to avoid errors' common to all MC programs?

[Hideki Kato]

Darren Cook: <4D3CF95E.2020507 at dcook.org>:
>> On Fri, Jan 21, 2011 at 08:31:04PM +0900, Darren Cook wrote:
>>> Dynamic komi. (E.g. if it thinks it has only 30% chance of winning at
>>> 7.5pt komi, but if you reduce the komi to 5.5pts it thinks it has a 55%
>>> chance of winning, then reduce the komi to 5.5pts: it will play an
>>> intelligent looking endgame and lose by 1.5pts.)
>>>
>>> This will make it weaker overall (because it won't try so hard to cause
>>> trouble)
>> 
>> What do you base this statement on? It's rather controversial for me. :-)
>
>Without dynamic komi the program will choose the move to maximize its
>winning rate. If dynamic komi causes a different move to be chosen then
>it implies it is choosing a move that it thinks has a less than or equal
>chance of winning.

No, unless the bot is perfect.

If an MC bot has some systematic bias in the simulations (a typical 
case is the nakade problem) then he has some chance to lose a game 
even if being ahead in the middle.  Since a bot with dynamic-komi can 
have more margin in the middle of a game, the chance to keep the lead 
and win the game could increase in such cases.

Another possibility is the case that opponent plays some mistakes 
(especially at near the end of a game).  An MC bot frequently plays 
hopeless and losing moves in a losing position and makes the behind 
bigger but dyanmic-komi could reduce this bad habit. 

Hideki

>Also, I believe Don did some (self-play?) experiments a few years ago
>and the dynamic komi version lost more; I think this is where Don's
>coolness to dynamic komi comes from (apologies if my memory is inaccurate).
>
>However, Magnus has some experiments with Valkyria where the dynamic
>komi version was stronger even in non-handicap self-play games.
>
>>> I.e. program endgame is generally stronger than the humans of the same
>>> rank; chances are a 1-dan human will make a few 1pt or 2pt errors during
>>> the endgame.
>> 
>> I also think this is not obviously true at all. My observations have
>> been that MCTS does not perform too well at all in very close endgames.
>> (Though it is not a big disadvantage in practice since it is in the
>> nature of MCTS to strive for deciding the game ASAP, i.e. in the middle
>> game.)
>
>My study has mostly been of 9x9 games, and as long as there is not a
>seki on the board the MCTS programs will practically never lose if they
>have a winning position at move 30. (Extra condition for that statement:
>Chinese rules.)
>
>Darren
-- 
Hideki Kato <mailto:hideki_katoh at ybb.ne.jp>