[Computer-go] Skip-opening matchmode

Darren Cook darren at dcook.org
Thu Jun 5 03:05:04 PDT 2014


> But the Bots are deluded by the Fata Morgana of huge moyos and will never
> get to such a position. We don't know how Bots perform on those positions,
> because they simply never happen. And such tests could actually show, if
> there is a potential improvement of the bots, when giving them some
> different starting position into the middle game.

Yes, this is a good argument [1]. Self-play, and play between bots all
using very similar algorithms, is claimed to be the best evaluation
method. If your goal is to be one stone stronger than other programs, no
argument from me. But beyond that it becomes misleading.

Darren

[1]: This was the same argument I made in favour of a very large test
suite (e.g. 10,000 positions, from all stages of the game, and all
styles and strengths of game), as the best way to evaluate progress.



-- 
Darren Cook, Software Researcher/Developer
My new book: Data Push Apps with HTML5 SSE
Published by O'Reilly: (ask me for a discount code!)
  http://shop.oreilly.com/product/0636920030928.do
Also on Amazon and at all good booksellers!



More information about the Computer-go mailing list