But I remember thinking that it might still be useful for evaluation. I remember thinking that that if you compared pseudo liberties to real liberties it might tell you something. If the ratio was high (a lot more pseudo liberties) it might indicate that the liberties are easy to protect. But there was a definite negative response to the idea that they were useful in any way.
- Don On Sun, 2007-05-27 at 23:54 -0400, Jason House wrote: > Don Dailey wrote: > > I remember that conversation and the negative response. But to be fair > > to the ones who were negative, you presented this as an evaluation > > feature that could be calculated quickly, not as a pure performance > > optimization. The negative response was in response to the suggestion > > that it might be used as an evaluation feature instead of true > > liberties. > > > > At least that is how I remember it. > > > > Oh, I definitely didn't mean to say or imply that those who gave a > negative reaction were wrong. I apologize if anyone took my post that > way! At the time, MC wasn't a big push in computer go, and I know I > wasn't thinking about rapid random games. You're absolutely right that > I was looking to use it as some kind of quick and dirty evaluation > feature. Honestly, I think the feedback was helpful too. I ended up > making a concept of local liberties that were slightly smarter than > pseudo liberties and were much more useful for non-MC computations. _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
