from pettingzoo. . Each game is fixed with two players, two rounds, two-bet maximum andraise amounts of 2 and 4 in the first and second round. . Training CFR (chance sampling) on Leduc Hold’em; Having Fun with Pretrained Leduc Model; Training DMC on Dou Dizhu; Evaluating Agents. A second related (offline) approach in-cludes counterfactual values for game states that could have been reached off the path to the endgames (Jackson 2014). Find hotels in Leduc from CA $61. {"payload":{"allShortcutsEnabled":false,"fileTree":{"":{"items":[{"name":"experiments","path":"experiments","contentType":"directory"},{"name":"models","path":"models. Leduc Hold'em은 Texas Hold'em의 단순화 된. We also evaluate SoG on the commonly used small benchmark poker game Leduc hold’em, and a custom-made small Scotland Yard map, where the approximation quality compared to the optimal policy can be computed exactly. . Each game is fixed with two players, two rounds, two-bet maximum and raise amounts of 2 and 4 in the first and second round. . DeepStack for Leduc Hold'em. to bridge reinforcement learning and imperfect information games. Jonathan Schaeffer. Now that we have a basic understanding of the structure of environment repositories, we can start thinking about the fun part - environment logic! For this tutorial, we will be creating a two-player game consisting of a prisoner, trying to escape, and a guard, trying to catch the prisoner. . Return type: (dict) rlcard. . This allows PettingZoo to represent any type of game multi-agent RL can consider. static judge_game (players, public_card) ¶ Judge the winner of the game. The goal of RLCard is to bridge reinforcement learning and imperfect information games, and push. . There are two rounds. Dirichlet distributions offer a simple prior for multinomi- 6 Experimental Setup als, which is a. 2. ''' A toy example of playing against pretrianed AI on Leduc Hold'em. For learning in Leduc Hold’em, we manually calibrated NFSP for a fully connected neural network with 1 hidden layer of 64 neurons and rectified linear. . The second round consists of a post-flop betting round after one board card is dealt. 2 2 Background 5 2. InfoSet Number: the number of the information sets; Avg. 3, bumped all versions. env() average_total_reward(env, max_episodes=100, max_steps=10000000000) Where max_episodes and max_steps both limit the total. Dou Dizhu (wiki, baike). This amounts to the first action abstraction algorithm (algo-rithm for selecting a small number of discrete actions to use from a continuum of actions—a key preprocessing step forPettingZoo’s API has a number of features and requirements. -Player with same card as op wins, else highest card. In this paper, we provide an overview of the key. in imperfect-information games, such as Leduc Hold’em (Southey et al. from rlcard. limit-holdem-rule-v1. Leduc Hold’em; Rock Paper Scissors; Texas Hold’em No Limit; Texas Hold’em; Tic Tac Toe; MPE. The interfaces are exactly the same to OpenAI Gym. {"payload":{"allShortcutsEnabled":false,"fileTree":{"":{"items":[{"name":"experiments","path":"experiments","contentType":"directory"},{"name":"models","path":"models. Leduc Hold’em (a simplified Te xas Hold’em game), Limit. '''. leducholdem_rule_models. Run examples/leduc_holdem_human. leduc-holdem. Waterworld is a simulation of archea navigating and trying to survive in their environment. md","contentType":"file"},{"name":"blackjack_dqn. 10^4. #. test import api_test from pettingzoo. You can try other environments as well. For a comparison with the AEC API, see About AEC. doc, example. Leduc Hold'em is a simplified version of Texas Hold'em. It is played with a deck of six cards, comprising two suits of three ranks each (often the king, queen, and jack - in our implementation, the ace, king, and queen). . . Discover the meaning of the Leduc name on Ancestry®. However, if their choices are different, the winner is determined as follows: rock beats scissors, scissors beat paper, and paper beats rock. DeepStack is an artificial intelligence agent designed by a joint team from the University of Alberta, Charles University, and Czech Technical University. py to play with the pre-trained Leduc Hold'em model:Leduc hold'em is a simplified version of texas hold'em with fewer rounds and a smaller deck. , 2011], both UCT-based methods initially learned faster than Outcome Sampling but UCT later suf-fered divergent behaviour and failure to converge to a Nash equilibrium. The deck used in UH-Leduc Hold’em, also call . 5 2 0 50 100 150 200 250 300 Exploitability Time in s XFP, 6-card Leduc FSP:FQI, 6-card Leduc Figure:Learning curves in Leduc Hold’em. 在Leduc Hold'em是双人游戏, 共有6张卡牌: J, Q, K各两张. The game begins with each player. 185, Section 5. Fig. Simple; Simple Adversary; Simple Crypto; Simple Push; Simple Reference; Simple Speaker Listener; Simple Spread; Simple Tag; Simple World Comm; SISL. Rule. Below is an example: from pettingzoo. Leduc Hold’em is a two player poker game. 2 Kuhn Poker and Leduc Hold’em. :param state: Raw state from the game :type. After training, run the provided code to watch your trained agent play vs itself. I am using the simplified version of Texas Holdem called Leduc Hold'em to start. This tutorial was created from LangChain’s documentation: Simulated Environment: PettingZoo. agents import RandomAgent. Advanced PPO: CleanRL’s official PPO example, with CLI, TensorBoard and WandB integration. Returns: list of payoffs. butterfly import pistonball_v6 env = pistonball_v6. It supports various card environments with easy-to-use interfaces, including Blackjack, Leduc Hold'em, Texas Hold'em, UNO, Dou Dizhu and Mahjong. We show that our proposed method can detect both assistant and associa-tion collusion. '>classic. The same to step. 为此，东京大学的研究人员引入了Suspicion Agent这一创新智能体，通过利用GPT-4的能力来执行不完全信息博弈。. Confirming the observations of [Ponsen et al. The most Leduc families were found in Canada in 1911. The first computer program to outplay human professionals at heads-up no-limit Hold'em poker. Abstract We present RLCard, an open-source toolkit for reinforce- ment learning research in card games. 4 with a fix for texas hold'em no limit; bump version; 1. in games with small decision space, such as Leduc hold’em and Kuhn Poker. Leduc hold'em for 2 players. . . At the end, the player with the best hand wins and. Rules can be found here. from rlcard. . No limit is placed on the size of the bets, although there is an overall limit to the total amount wagered in each game ( 10 ). . 游戏过程很简单, 首先, 两名玩家各投1个筹码作为底注(也有大小盲玩法, 即一个玩家下1个筹码, 另一个玩家下2个筹码). Both variants have a small set of possible cards and limited bets. State Representation of Blackjack; Action Encoding of Blackjack; Payoff of Blackjack; Leduc Hold’em. Raw Blame. Head coach Michael LeDuc of Damien hugs his wife after defeating Clovis North 65-57 to win the CIF State Division I boys basketball state championship game at Golden 1 Center in Sacramento on. static step (state) ¶ Predict the action when given raw state. Toggle navigation of MPE. Leduc Hold'em as Single-Agent Environment. 4. Leduc Hold'em is a common benchmark in imperfect-information game solving because it is small enough to be solved but still. It supports various card environments with easy-to-use interfaces, including Blackjack, Leduc Hold'em, Texas Hold'em, UNO, Dou Dizhu and Mahjong. 10^0. The Control Panel provides functionalities to control the replay process, such as pausing, moving forward, moving backward and speed control. games: Leduc Hold’em [Southey et al. However, if their choices are different, the winner is determined as follows: rock beats scissors, scissors beat paper, and paper beats rock. The winner will receive +1 as a reward and the loser will get -1. In addition, we also prove that the weighted average strategy by skipping previous itera- The most popular variant of poker today is Texas hold’em. cfr --cfr_algorithm external --game Leduc. RLCard 提供人机对战 demo。RLCard 提供 Leduc Hold'em 游戏环境的一个预训练模型，可以直接测试人机对战。Leduc Hold'em 是一个简化版的德州扑克，游戏使用 6 张牌（红桃 J、Q、K，黑桃 J、Q、K），牌型大小比较中 对牌>单牌，K>Q>J，目标是赢得更多的筹码。Poker and Leduc Hold’em. . Toggle navigation of MPE. game - this file defines that we are playing the game of Leduc hold'em. In the rst round a single private card is dealt to each. , 2005] and Flop Hold’em Poker (FHP) [Brown et al. . For learning in Leduc Hold’em, we manually calibrated NFSP for a fully connected neural network with 1 hidden layer of 64 neurons and rectified linear activations. In the rst round a single private card is dealt to each. Leduc Hold ‘em Rule agent version 1. 10^48. . Leduc Hold'em is a smaller version of Limit Texas Hold'em (first introduced in Bayes' Bluff: Opponent Modeling in Poker). This allows PettingZoo to represent any type of game multi-agent RL can consider. After training, run the provided code to watch your trained agent play vs itself. 2: The 18 Card UH-Leduc-Hold’em Poker Deck. Read writing from Ziad SALLOUM on Medium. env = rlcard. . RLCard provides unified interfaces for seven popular card games, including Blackjack, Leduc Hold’em (a simplified Texas Hold’em game), Limit Texas Hold’em, No-Limit. , Queen of Spade is larger than Jack of. In Leduc hold ’em, the deck consists of two suits with three cards in each suit. Leduc Hold’em 10 210 100 Limit Texas Hold’em 1014 103 100 Dou Dizhu 1053 ˘1083 1023 104 Mahjong 10121 1048 102 No-limit Texas Hold’em 10162 103 104 UNO 10163 1010 101 Table 1: A summary of the games in RLCard. In order to encourage and foster deeper insights within the community, we make our game-related data publicly available. The AEC API supports sequential turn based environments, while the Parallel API. At the beginning, both players get two cards. . , 2011], both UCT-based methods initially learned faster than Outcome Sampling but UCT later suf-fered divergent behaviour and failure to converge to a Nash equilibrium. It uses pure PyTorch and is written in only ~4000 lines of code. The experiment results demonstrate that our algorithm significantly outperforms NE baselines against non-NE opponents and keeps low exploitability at the same time. Note that this library is intended to. 5. , 2005] and Flop Hold’em Poker (FHP) [Brown et al. 2 and 4), at most one bet and one raise. ,2017]techniques to automatically construct different collusive strategies for both environments. . 01 every time they touch an evader. 🤖 An Open Source Texas Hold'em AI Topics. Having fun with pretrained Leduc model; Leduc Hold'em as single-agent environment; Training CFR on Leduc Hold'em; Demo. 0# Released on 2021-08-02 - GitHub - PyPI-Upgraded to RLCard 1. using two different heads-up limit poker variations: a small-scale variation called Leduc Hold’em, and a full-scale one called Texas Hold’em. Moreover, RLCard supports flexible en viron- Leduc Hold’em. py. , 2019]. , 2005) and Flop Hold’em Poker (FHP)(Brown et al. Leduc Hold’em : 10^2: 10^2: 10^0: leduc-holdem: doc, example: Limit Texas Hold'em (wiki, baike) 10^14: 10^3: 10^0: limit-holdem: doc, example: Dou Dizhu (wiki, baike) 10^53 ~ 10^83: 10^23: 10^4: doudizhu: doc, example: Mahjong (wiki, baike) 10^121: 10^48: 10^2: mahjong: doc, example: No-limit Texas Hold'em (wiki, baike) 10^162: 10^3: 10^4: no. He has always been there toReinforcement Learning / AI Bots in Card (Poker) Games - Blackjack, Leduc, Texas, DouDizhu, Mahjong, UNO. . , 2011], both UCT-based methods initially learned faster than Outcome Sampling but UCT later suf-fered divergent behaviour and failure to converge to a Nash equilibrium. . Leduc Hold’em : 10^2: 10^2: 10^0: leduc-holdem: doc, example: Limit Texas Hold'em (wiki, baike) 10^14: 10^3: 10^0: limit-holdem: doc, example: Dou Dizhu (wiki, baike) 10^53 ~ 10^83: 10^23: 10^4: doudizhu: doc, example: Mahjong (wiki, baike) 10^121: 10^48: 10^2: mahjong: doc, example: No-limit Texas Hold'em (wiki, baike) 10^162: 10^3: 10^4: no. In the example, there are 3 steps to build an AI for Leduc Hold’em. Parameters: players (list) – The list of players who play the game. In addition to NFSP’s main, average strategy profile we also evaluated the best response and greedy-average strategies, which deterministically choose actions that maximise the predicted ac- tion values or probabilities respectively. This environment is part of the classic environments. At the beginning, both players get two cards. Leduc Hold'em is a smaller version of Limit Texas Hold'em (first introduced in Bayes' Bluff: Opponent Modeling in Poker). CleanRL is a lightweight,. Cooperative pong is a game of simple pong, where the objective is to keep the ball in play for the longest time. Leduc Formation, a stratigraphical unit in the Western Canadian Sedimentary Basin. Written by Thomas Trenner. Artificial Intelligence----Follow. ipynb","path. Every time the pursuers fully surround an evader each of the surrounding agents receives a reward of 5 and the evader is removed from the environment. Simple; Simple Adversary; Simple Crypto; Simple Push;. 0. If you look at pg. . 游戏过程很简单, 首先, 两名玩家各投1个筹码作为底注(也有大小盲玩法, 即一个玩家下1个筹码, 另一个玩家下2个筹码). The results show that Suspicion-Agent can potentially outperform traditional algorithms designed for imperfect information games, without any specialized training or examples. Reinforcement Learning / AI Bots in Get Away. Training CFR (chance sampling) on Leduc Hold'em; Having fun with pretrained Leduc model; Leduc Hold'em as single-agent environment; R examples can be found here. Pursuers also receive a reward of 0. utils import TerminateIllegalWrapper env = OpenSpielCompatibilityV0(game_name="chess", render_mode=None) env = TerminateIllegalWrapper(env, illegal_reward=-1) env. The results show that Suspicion-Agent can potentially outperform traditional algorithms designed for imperfect information games, without any specialized. Only player 2 can raise a raise. The goal of RLCard is to bridge reinforcement learning and imperfect information games, and push forward the research of reinforcement learning in domains with multiple agents, large. . Leduc Hold’em is a poker variant that is similar to Texas Hold’em, which is a game often used in academic research . agent_iter(): observation, reward, termination, truncation, info = env. The goal of this thesis work is the design, implementation, and evaluation of an intelligent agent for UH Leduc Poker. Fig. Here is a definition taken from DeepStack-Leduc. Leduc Hold'em is a simplified version of Texas Hold'em. Run examples/leduc_holdem_human. Rule-based model for Limit Texas Hold’em, v1. RLCard is an open-source toolkit for reinforcement learning research in card games. 59 KB. py at master · datamllab/rlcard# These arguments are fixed in Leduc Hold'em Game # Raise amount and allowed times: self. The ε-greedy policies’ exploration started at 0. We can know that the Leduc Hold'em environment is a 2-player game with 4 possible actions. in imperfect-information games, such as Leduc Hold’em (Southey et al. But unlike in Limit Texas Hold'em game in which each player can only choose a fixed amount of raise and the number of raises is limited. . Special UH-Leduc-Hold’em Poker Betting Rules: Ante is $1, raises are exactly $3. 1 Adaptive (Exploitative) Approach. The deck consists only two pairs of King, Queen and Jack, six cards in total. Note you can easily find yourself in a dead-end escapable only through the. Rules can be found here. ,2012) when compared to established methods like CFR (Zinkevich et al. 10^2. both Texas and Leduc hold’em, using two different classes of priors: independent Dirichlet and an informed prior pro-vided by an expert. At the beginning of a hand, each player pays a one chip ante to. RLCard is an open-source toolkit for reinforcement learning research in card games. . . We demonstrate the effectiveness of this technique in Leduc Hold'em against opponents that use the UCT Monte Carlo tree search algorithm. Training CFR (chance sampling) on Leduc Hold’em¶ To show how we can use step and step_back to traverse the game tree, we provide an example of solving Leduc Hold’em with CFR (chance sampling). AI. @article{terry2021pettingzoo, title={Pettingzoo: Gym for multi-agent reinforcement learning}, author={Terry, J and Black, Benjamin and Grammel, Nathaniel and Jayakumar, Mario and Hari, Ananth and Sullivan, Ryan and Santos, Luis S and Dieffendahl, Clemens and Horsch, Caroline and Perez-Vicente, Rodrigo and others}, journal={Advances in Neural. >> Leduc Hold'em pre-trained model >> Start a new game! >> Agent 1 chooses raise. in imperfect-information games, such as Leduc Hold’em (Southey et al. . py. Toggle navigation of MPE. . Find your family's origin in Canada, average life expectancy, most common occupation, and. This size is two chips in the first betting round and four chips in the second. The second round consists of a post-flop betting round after one board card is dealt. Creator of Every day, Ziad SALLOUM and thousands of other voices read, write, and share important stories on Medium. 120 lines (98 sloc) 3. Rules can be found here. So that good agents. This tutorial is made with two target audiences in mind: (1) Those with an interest in poker who want to understand how AI. Moreover, RLCard supports flexible environ-in Leduc hold’em (top left), goofspiel (top center), and random goofspiel (top right). Toggle navigation of MPE. py","path":"tutorials/Ray/render_rllib_leduc_holdem. Leduc Hold’em; Rock Paper Scissors; Texas Hold’em No Limit; Texas Hold’em; Tic Tac Toe; MPE. Neural Networks. In 1840 there were 3. . approach. Conversion wrappers# AEC to Parallel#. Leduc Hold ‘em rule model. , 2019]. RLlib is an industry-grade open-source reinforcement learning library. uno-rule-v1. 52 KB. Leduc Hold ‘em Rule agent version 1. The stages consist of a series of three cards ("the flop"), later an additional single card ("the. . UH-Leduc Hold’em Deck: This is a “ queeny ” 18-card deck from which we draw the players’ card sand the flop without replacement. The library currently implements vanilla CFR [1], Chance Sampling (CS) CFR [1,2], Outcome Sampling (CS) CFR [2], and Public Chance Sampling (PCS) CFR [3]. The latter is a smaller version of Limit Texas Hold’em and it was introduced in the research paper Bayes’ Bluff: Opponent Modeling in Poker in 2012. Sequence-form. Firstly, tell “rlcard” that we need a Leduc Hold’em environment. The deck used in Leduc Hold’em contains six cards, two jacks, two queens and two kings, and is shuffled prior to playing a hand. . . In the rst round a single private card is dealt to each. chisness / leduc2. >> Leduc Hold'em pre-trained model >> Start a new game! >> Agent 1 chooses raise. . Solve Leduc Hold Em using cfr. . md","contentType":"file"},{"name":"best_response. . In Leduc hold ’em, the deck consists of two suits with three cards in each suit. Apart from rule-based collusion, we use Deep Reinforcement Learning (Arulkumaran et al. See the documentation for more information. The goal of RLCard is to bridge reinforcement learning and imperfect information games, and push forward the research of reinforcement learning in domains with mul-tiple agents, large state and action space, and sparse reward. Training CFR on Leduc Hold'em. py to play with the pre-trained Leduc Hold'em model. doudizhu. Next time, we will finally get to look at the simplest known Hold’em variant, called Leduc Hold’em, where a community card is being dealt between the first and second betting rounds. To show how we can use step and step_back to traverse the game tree, we provide an example of solving Leduc Hold'em with CFR (chance sampling). 10^0. Pre-trained CFR (chance sampling) model on Leduc Hold’em. We evaluate SoG on four games: chess, Go, heads-up no-limit Texas hold’em poker, and Scotland Yard. g. . We present experiments in no-limit Leduc Hold’em and no-limit Texas Hold’em to optimize bet sizing. . 2017) tech-niques to automatically construct different collusive strate-gies for both environments. Note that for both . Example implementation of the DeepStack algorithm for no-limit Leduc poker - GitHub - matthewmav/MIB: Example implementation of the DeepStack algorithm for no-limit Leduc pokerLeduc Hold’em; Rock Paper Scissors; Texas Hold’em No Limit; Texas Hold’em; Tic Tac Toe; MPE. class rlcard. Many classic environments have illegal moves in the action space. Parameters: players (list) – The list of players who play the game. A few years back, we released a simple open-source CFR implementation for a tiny toy poker game called Leduc hold'em link. Demo. leduc-holdem-rule-v1. Mahjong (wiki, baike) 10^121. It supports various card environments with easy-to-use interfaces, including Blackjack, Leduc Hold'em, Texas Hold'em, UNO, Dou Dizhu and Mahjong. . We will walk through the creation of a simple Rock-Paper-Scissors environment, with example code for both AEC and Parallel environments. Apart from rule-based collusion, we use Deep Re-inforcementLearning[Arulkumaranetal. Leduc Hold ’Em. ''' A toy example of playing against pretrianed AI on Leduc Hold'em. CleanRL Tutorial#. Training CFR on Leduc Hold'em; Having fun with pretrained Leduc model; Leduc Hold'em as single-agent environment; R examples can be found here. A solution to the smaller abstract game can be computed and isReinforcement Learning / AI Bots in Card (Poker) Game: New limit Holdem - GitHub - gsiatras/Reinforcement_Learning-Q-learning_and_Policy_Iteration_Rlcard. Leduc Hold'em is a simplified version of Texas Hold'em. cfr --cfr_algorithm external --game Leduc. . Contribute to mjiang9/_rlcard development by creating an account on GitHub. We release all interaction data between Suspicion-Agent and traditional algorithms for imperfect-informationTraining CFR on Leduc Hold'em In this tutorial, we will showcase a more advanced algorithm CFR, which uses step and step_back to traverse the game tree. We release all interaction data between Suspicion-Agent and traditional algorithms for imperfect-informationin imperfect-information games, such as Leduc Hold’em (Southey et al. Leduc Hold’em and River poker. Also added support for num_players in RLcard based environments which can have variable numbers of players. Figure 2: Visualization modules in RLCard of Dou Dizhu (left) and Leduc Hold’em (right) for algorithm debugging. . . In a Texas Hold’em game, just from the first round alone, we move from 52c2*50c2 = 1,624,350 to 28,561 combinations by using lossless abstraction. Test your understanding by implementing CFR (or CFR+ / CFR-D) to solve one of these two games in your favorite programming language. . Each player can only check once and raise once; in the case a player is not allowed to check . Rule. Toggle navigation of MPE. Leduc Hold'em. It supports various card environments with easy-to-use interfaces, including Blackjack, Leduc Hold'em, Texas Hold'em, UNO, Dou Dizhu and Mahjong. Training CFR (chance sampling) on Leduc Hold’em¶ To show how we can use step and step_back to traverse the game tree, we provide an example of solving Leduc Hold’em with CFR (chance sampling). When it is played with just two players (heads-up) and with fixed bet sizes and a fixed number of raises (limit), it is called heads-up limit hold’em or HULHE ( 19 ). We present experiments in no-limit Leduc Hold’em and no-limit Texas Hold’em to optimize bet sizing. md","path":"docs/README. leduc-holdem-cfr. 75 times the size of the pursuer radius, while food. . {"payload":{"allShortcutsEnabled":false,"fileTree":{"":{"items":[{"name":"README. Most environments only give rewards at the end of the games once an agent wins or losses, with a reward of 1 for winning and -1 for losing. We have also constructed a smaller version of hold ’em, which seeks to retain the strategic ele-ments of the large game while keeping the size of the game tractable. Jonathan Schaeffer. 3. Leduc Hold’em 10^2 10^2 10^0 leduc-holdem 文档, 释例 限注德州扑克 Limit Texas Hold'em (wiki, 百科) 10^14 10^3 10^0 limit-holdem 文档, 释例 斗地主 Dou Dizhu (wiki, 百科) 10^53 ~ 10^83 10^23 10^4 doudizhu 文档, 释例 麻将 Mahjong (wiki, 百科) 10^121 10^48 10^2 mahjong 文档, 释例Leduc Hold’em (a simplified Texas Hold’em game), Limit Texas Hold’em, No-Limit Texas Hold’em, UNO, Dou Dizhu and Mahjong. In PettingZoo, we can use action masking to prevent invalid actions from being taken. The game is over when the ball goes out of bounds from either the left or right edge of the screen. #. ↳ 15 cells hiddenThe following script uses pytest to test all other PettingZoo environments which support action masking. 1. ,2012) when compared to established methods like CFR (Zinkevich et al. It supports various card environments with easy-to-use interfaces, including Blackjack, Leduc Hold’em, Texas Hold’em, UNO, Dou Dizhu and Mahjong. We have also constructed a smaller version of hold ’em, which seeks to retain the strategic ele-ments of the large game while keeping the size of the game tractable. A simple rule-based AI. , 2011], both UCT-based methods initially learned faster than Outcome Sampling but UCT later suf-fered divergent behaviour and failure to converge to a Nash equilibrium. leduc-holdem-cfr. Leduc Hold'em에서 CFR 교육; 사전 훈련 된 Leduc 모델로 즐거운 시간 보내기; 단일 에이전트 환경으로서의 Leduc Hold'em; R 예제는 여기 에서 찾을 수 있습니다. Leduc Hold ’Em. After training, run the provided code to watch your trained agent play. 1 Adaptive (Exploitative) Approach. This project used two types of reinforcement learning (SARSA and Q-Learning) to train agents to play a modified version of Leduc Hold'em Poker.