A playout just returs score and moves with colors in my implementation. -Hideki
David Fotland: <[EMAIL PROTECTED]>: >Helps a little, but I'm still unclear on how a new node is handled. > > > >Do you have lightweight nodes (that only contain info on one positions, with >pointers to children and siblings)? Then when uct finds a node with no >children, what do you do? Mogo and Mark create a new node for each new >simulation. So do you create a single random child at this point and >continue, or do you create all children, then pick one, and continue? > > > >The descriptions of RAVE are also pretty unclear. I guess each node has a >rave counter that counts how many times this move was seen during a playout. >When you backup data in the uct tree, do you visit every child and update >the rave counters of the children? > > > >David > > > >From: [EMAIL PROTECTED] >[mailto:[EMAIL PROTECTED] On Behalf Of Jason House >Sent: Tuesday, February 05, 2008 11:15 AM >To: computer-go >Subject: Re: [computer-go] More UCT / Monte-Carlo questions > > > > > >On Feb 5, 2008 2:08 PM, David Fotland <[EMAIL PROTECTED]> wrote: > >It think many programs run several simulations through a node before >allocating the children. I can see how this saves memory, but then how do >you save the RAVE information from the early simulations? > > >For RAVE, after doing a sim at a leaf, I walk back up the UCT tree. At each >internal node, I keep update RAVE game counters applicable children. I >never go deeper into the tree to apply RAVE values. Does that help? > > >---- inline file >_______________________________________________ >computer-go mailing list >[email protected] >http://www.computer-go.org/mailman/listinfo/computer-go/ -- [EMAIL PROTECTED] (Kato) _______________________________________________ computer-go mailing list [email protected] http://www.computer-go.org/mailman/listinfo/computer-go/
