Petr Baudis wrote:
I forgot to ask about another unclear thing in your paper - you say that
in simulations, you use the continuity feature. Does that effectively
mean distance-to-last-move feature with just two levels, one for d=3,
another for d3?
Yes. I believe I called it contiguity.
Rémi
Petr Baudis wrote:
On Wed, Dec 16, 2009 at 08:57:24PM +0100, Rémi Coulom wrote:
The problem is that I am not using the same set of features for
biasing the tree, and for playouts. Playouts only use fast light
features. The tree part uses slow complex features. In particular, I
use patterns of
Petr Baudis wrote:
On Wed, Dec 16, 2009 at 08:57:24PM +0100, Rémi Coulom wrote:
The problem is that I am not using the same set of features for
biasing the tree, and for playouts. Playouts only use fast light
features. The tree part uses slow complex features. In particular, I
use patterns of
On Sat, Jan 02, 2010 at 01:49:06PM +0100, Petr Baudis wrote:
On Wed, Dec 16, 2009 at 08:57:24PM +0100, Rémi Coulom wrote:
The problem is that I am not using the same set of features for
biasing the tree, and for playouts. Playouts only use fast light
features. The tree part uses slow
You say radius = 3, then 3x3 patterns. Which is it? Radius 3 would be 5x5
to 7x7, depending on how you define the radius.
David
-Original Message-
From: computer-go-boun...@computer-go.org [mailto:computer-go-
boun...@computer-go.org] On Behalf Of Petr Baudis
Sent: Saturday,
On Sat, Jan 02, 2010 at 09:58:10AM -0800, David Fotland wrote:
You say radius = 3, then 3x3 patterns. Which is it? Radius 3 would be 5x5
to 7x7, depending on how you define the radius.
I use it in the CrazyStone/Stern sense, where they use what I call the
gridcular metric: d = dx + dy +