Image Image Image




Post new topic Reply to topic  [ 3 posts ] 
Author Message
 Post subject: Slumbot: An Implementation Of Counterfactual Regret Minimization
PostPosted: Sun Aug 12, 2012 2:13 pm 
Offline
Senior member
User avatar

Posts: 124
Favourite Bot: coming
Slumbot: An Implementation Of Counterfactual Regret Minimization On Commodity Hardware

Authors : Eric Jackson

Abstract :

This paper describes the design and implementation of Slumbot, a competitor in the Annual Computer Poker Competition in the heads-up limit division for the last few years.
My approach this year can be viewed as an implementation of the Public Chance Sampling (PCS) variant of Counterfactual Regret Minimization (CFR) (see [1], [2]).
However, Slumbot differs from some implementations in the way that it manages to utilize a very large abstraction while still running on low-end commodity hardware.
The 2011 version of Slumbot finished second in the equilibrium competition for heads-up limit. The 2012 version appears to beat the 2011 version by around 12 milli-big-blinds per hand.

http://www.ualberta.ca/~archibal/papers/jackson.pdf


Top
 Profile E-mail  
 
 Post subject: Re: Slumbot: An Implementation Of Counterfactual Regret Minimization
PostPosted: Sun Aug 19, 2012 1:14 pm 
Offline
PokerAI fellow
User avatar

Posts: 1115
Favourite Bot: Johnny #5
Wow, that is interesting... I give him credit for a radical implementation. It blows my mind that 1440 iterations were enough to not only be competitive, but win first place.


Top
 Profile  
 
 Post subject: Re: Slumbot: An Implementation Of Counterfactual Regret Minimization
PostPosted: Sun Aug 19, 2012 8:46 pm 
Offline
Regular member
User avatar

Posts: 50
Favourite Bot: wut
c2008 wrote:
Wow, that is interesting... I give him credit for a radical implementation. It blows my mind that 1440 iterations were enough to not only be competitive, but win first place.


well, it's not vanilla iterations. "In my particular implementation, I fully explore all the possible flop
and turn boards on each iteration, but sample only a single river."


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 3 posts ] 


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: