[Computer-go] evaluating number of wins versus looses
pasky at ucw.cz
Mon Mar 30 11:40:32 PDT 2015
On Mon, Mar 30, 2015 at 04:17:13PM +0200, Erik van der Werf wrote:
> On Mon, Mar 30, 2015 at 4:09 PM, Petr Baudis <pasky at ucw.cz> wrote:
> > The strongest programs often use RAVE or LGRF or something like that,
> > with or without the UCB for tree exploration.
> Huh, are there any strong programs that got LGRF to work?
Oops, I actually meant to write criticality, though I'm a little bit
confused now whether Nomitan primarily uses criticality, LGRF or both.
Perhaps Simon will want to clarify...
(In general, it seems apparent that many methods of information sharing
are complementary to a large degree, i.e. by opting for one at the
beginning, you get similar benefit as if you did a different one,
especially if you tune your way to a good local optimum. But RAVE
If you do not work on an important problem, it's unlikely
you'll do important work. -- R. Hamming
More information about the Computer-go