What about the komi discussion, now that KataGo is coming

For discussing go computing, software announcements, etc.
hyperpape
Tengen
Posts: 4382
Joined: Thu May 06, 2010 3:24 pm
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 4k
Location: Caldas da Rainha, Portugal
Has thanked: 499 times
Been thanked: 727 times

Re: What about the komi discussion, now that KataGo is comin

Post by hyperpape »

moha wrote:Otherwise common sense would predict 7.5 is worse: correct komi is 7 and it is better to let B win ties than W, since practical players are not close to correct/perfect play thus need slightly less komi.
I've long been cautious about this assumption. It certainly is possible, but I'm not sure that anyone has demonstrated that it's true.

Reading this, it just occurred to me to wonder if we could analyze training data for AI to test this idea.
Post Reply