Articles
Theory explained
Browse the full library of GTO Hero articles.
Monte Carlo CFR - How to Make a Poker AI Part 4
Monte Carlo refers to an algorithm that utilises randomisation to gain a sample that is an accurate representation of the full population. The name refers to theCasino de Monte-Carlo; a legendary icon of gambling.
Vanilla CFR Code and Results - How to make a poker AI part 3
Time to actually build our AI; we are going to build a CFR algorithm to find a Nash equilibrium strategy for heads up (2 players) No Limit Texas Hold 'em poker, in post flop scenarios. In this article I walk through the code that is related to the algorithm, run a simple scenario solve and then look at the results.
Vanilla CFR - How to Make a Poker AI Part 2
Vanilla CFR (Counterfactual Regret Minimisation) is the simplest form of CFR, it is the bare bones algorithm which we will build upon in future articles. In this article we will follow the theory needed for this algorithm to give us a Nash equilibrium strategy.
Introduction to Game Theory - How to make a poker AI part 1
Welcome to the first article in the "How to build a poker A.I" series. This series will delve into various aspects involved in building a poker A.I. The first few articles will explore the theory and algorithms behind solving games like poker; namely variants of CFR (Counter Factual Regret minimization) and how we can bring the impractically large game of poker down into an abstracted solvable size, we will then compare the algorithms we have built to see which converge quickest. But before doing all that, we first need to establish some basics! This article is written to give you the basics in game theory to help you understand future articles in this series.