[Pachi] A little bit lost in the code

Petr Baudis pasky at ucw.cz
Sat Dec 31 15:55:50 CET 2011


On Sat, Dec 31, 2011 at 12:49:35PM +0100, Jean-m. a. wrote:
> Hello. I wanted to try to implement an improved UCB algorithm
> (article UCB revisited: Improved regret bounds for the stochastic
> multi-armed bandit problem),
> but i'am a little bit lost in the code. where is the main part where the
> ucb algorithm is implemented.
> I read Petr Baudis' Master's Thesis <http://pasky.or.cz/go/prace.pdf>, and
> understand that the main policy of pachi is RAVE, but that
> you implemented ucb to , but can't find the code !

  The tree policy modules are in the uct/policy/ subdirectory.
ucb1amaf.c is the RAVE policy and ucb1.c is the classic plain
UCB1 policy.

  Have you tried reading the HACKING file? If you find it
incomprehensible, I would appreciate feedback. :-)


				Petr "Pasky" Baudis
	The goal of Computer Science is to build something that will
	last at least until we've finished building it.

More information about the Pachi mailing list