I have not read the papaer. Having said that... Rave cannot ruin the convergance guarantee because as the number of traversals through a node increases, the rave contribution decreases.
Virtual loss cannot ruin the convergence guarantee because it encourages exploration and there are a finite number of nodes to explore. Of course, I could be missing something. On Tue, Jun 7, 2011 at 6:59 PM, Darren Cook <[email protected]> wrote: > > The math escapes me here. I think doubling the playouts gains in the > > neighborhood of 70 ELO points. If adding a thread costs 10 ELO, > > adding more threads would stop being beneficial after about 14 > > threads. Doubling from 7 to 14 would lose 7*10 ELO, equaling the gain > > of the extra playouts. After that, adding threads should actually > > lose ELO. Yet we see people trying to put together systems with 100s > > of CPUs. What am I missing? > > Richard Segal (who operates Blue Fuego) has a paper on the upper limit > for scaling: > http://www.springerlink.com/content/b8p81h40129116kl/ > (Sorry, I couldn't find an author's download link for the paper; Richard > is on the Fuego list but I'm not sure he is even a lurker here.) > > I didn't fully understand the methodology, but what I did take away from > it (and discussions with Richard) was that though we're satisfied that > pure UCT eventually expands all nodes and can solve a position just like > minimax, this is not the case once you start adding enhancements such as > rave and virtual loss and *parallelizing the algorithm*. > > Darren > > > -- > Darren Cook, Software Researcher/Developer > > http://dcook.org/work/ (About me and my work) > http://dcook.org/blogs.html (My blogs and articles) > _______________________________________________ > Computer-go mailing list > [email protected] > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go >
_______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
