HackerPapers

arXiv:1407.5042cs.GT

Solving Large Imperfect Information Games Using CFR+

Oskari Tammelin

Early discovery. Linked by one identifiable commenter. This paper is outside the main ranking; a mention is not an endorsement.

Abstract

Counterfactual Regret Minimization and variants (e.g. Public Chance Sampling CFR and Pure CFR) have been known as the best approaches for creating approximate Nash equilibrium solutions for imperfect information games such as poker. This paper introduces CFR$^+$, a new algorithm that typically outperforms the previously known algorithms by an order of magnitude or more in terms of computation time while also potentially requiring less memory.

arXiv comments
Journal reference
DOI
Subjects
cs.GTComputer Science and Game Theory

On Hacker News

Linked by 1 identifiable commenter in 3 comments across 3 threads, from Jan 2017 to Sep 2021.

Comments linking this paper, per year.

Discussed in comments3 threads

  1. Bespoke Synth 1.0 – open-source software modular synthesizer 1 comment

    • > It was a shame there was the whole thing of losing the source code Yes, I can remember that very well! Eventually Oskari rebuilt the app basically from scratch, using .NET (at least IIRC) and that's what you get today if you go to the site and download it. It's not the…

      mastaziView on HNParent context

  2. DeepStack: Expert-Level Artificial Intelligence In Heads-Up No-Limit Poker 1 comment

    • With regards to Texas Hold'em, Conterfactual Regret Optimisation [1][2] was already able to beat humans in limit games, while it seems that this new approach is able to do the same in no-limit games. My naive understanding is that no-limit games are harder to crack because there…

      mastaziView on HNParent context

  3. Darpa Goes “Meta” with Machine Learning for Machine Learning (2016) 1 comment

    • This is completely OT but I just googled about the topic of "solving" Texas Hold'em Poker and other imperfect information games and discovered that one of the contributors to the research on Counterfactual Regret Minimization algorithms, developed at the University of…

      mastaziView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.