[computer-go] On expanding the UCT tree

[Heikki Levanto]

On Tue, May 01, 2007 at 11:42:46PM -0400, Chris Fant wrote:
> [...] memory-saving technique of not expanding a leaf until you have
> been through it many times (100, for example).

I have been wondering about this: If it pays off not to expand a node
until it has been visited 100 times, why not bite the bullet and make
those 100 simulations in one go? That should save a bit of time
traversing the tree up and down. Of course, it means that they all do
get a full 100 simulations, even if the first 90 show a bad result.
But it would make it easier to distribute the job to another thread,
processor, or even another computer.

-H

-- 
Heikki Levanto   "In Murphy We Turst"     heikki (at) lsd (dot) dk