Image Image Image




Post new topic Reply to topic  [ 1 post ] 
Author Message
 Post subject: CFRM - Dynamically Sized Bets
PostPosted: Sun Jan 20, 2013 8:52 am 
Offline
PokerAI fellow
User avatar

Posts: 1115
Favourite Bot: Johnny #5
Has anybody worked with this?

Thinking about it, it's just as simple as keeping track of the pot size and sampling probabilistically down a certain path, no? The probability for which (bet size) path to take is updated by the regret and normalized with some constant?

They discuss this bet-size probability update (formula 10) in the paper Automated Action Abstraction of Imperfect Information Extensive-Form Games. Though, I'm still not entirely sure what they're doing.

The game tree branching would remain f/c/r/a, where r is the dynamically sized bet node (assuming only one additional bet besides all-in). The node size in memory would remain the same (compared to a static abstraction), with the exception of an additional array (to store the bet regret probabilities). Makes sense, right?


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 1 post ] 


Who is online

Users browsing this forum: Google and 13 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: