InArtificial Intelligence in Plain EnglishbyThomas Trenner·May 24, 2020Building a Poker AI Part 8: Leduc Hold’em and a more generic CFR algorithm in PythonA CFR algorithm that works with arbitrary sequential games. We use it to compute a perfect strategy for Leduc Hold’em, a simple Poker game.A response icon3A response icon3
InArtificial Intelligence in Plain EnglishbyThomas Trenner·May 1, 2020Building a Poker AI Part 7: Exploitability, Multiplayer CFR and 3-player Kuhn PokerExploitability to measure the quality of our game-playing AI, multiplayer Counterfactual Regret Minimization and 3-player Kuhn PokerA response icon2A response icon2
InArtificial Intelligence in Plain EnglishbyThomas Trenner·Apr 18, 2020Building a Poker AI Part 6: Beating Kuhn Poker with CFR using PythonTrain an AI to learn an optimal strategy for Kuhn Poker, using an algorithm that learns through self-play.A response icon3A response icon3
InArtificial Intelligence in Plain EnglishbyThomas Trenner·Apr 13, 2020Building a Poker AI — Part 5: Sequential Games, Kuhn Poker and Counterfactual RegretsWe look at Kuhn Poker, the simplest Poker-style game, and explore how to extend regrets and regret matching to solve these types of games.
InArtificial Intelligence in Plain EnglishbyThomas Trenner·Apr 7, 2020Steps to building a Poker AI — Part 4: Regret Matching for Rock-Paper-Scissors in PythonPart 4 of my series on building a poker AI. We will develop the regret-matching algorithm in Python and apply it to Rock-Paper-Scissors.A response icon3A response icon3
InArtificial Intelligence in Plain EnglishbyThomas Trenner·Apr 5, 2020Steps to building a Poker AI — Part 3: Regrets and Minimizing Regrets in One-Shot GamesIntroducing the concept of regrets and a self-play algorithm to minimize expected future regrets through regret matching
InArtificial Intelligence in Plain EnglishbyThomas Trenner·Apr 3, 2020Steps to building a Poker AI — Part 2: Modelling Imperfect Information GamesPart 2 of my series on building a poker AI, explaining how to model games with hidden information, using Rock-Paper-Scissors as an example.
InArtificial Intelligence in Plain EnglishbyThomas Trenner·Apr 1, 2020Steps to building a Poker AI — Part 1: Outline and HistoryThis is the first in a series of articles that will outline the basic theory behind incomplete-information games (Poker being just one…