|
From: | Øystein Johansen |
Subject: | Re: [Bug-gnubg] Measuring Complexity |
Date: | Thu, 03 Sep 2009 21:37:31 +0200 |
User-agent: | Thunderbird 2.0.0.23 (Windows/20090812) |
Joseph Heled wrote: > I think it is not an easy one. My personal view is that a position is > "complex" if there are several *different* top moves which are very > close in equity. Now it all hangs on what "different" means :) Maybe we can measure "different" as the cosine to the angle between the neural net input vectors for the resulting positions? (just throwing an idea out in the air. I have no clue if it will work?) -Øystein
[Prev in Thread] | Current Thread | [Next in Thread] |