Texas Hold’em is a poker game involving 2 players and a regular 52 cards deck. At the beginning, both players get two cards. After betting, three community cards are shown and another round follows. At any time, a player could fold and the game will end. The winner will receive +1 as a reward and the loser will get -1. This is an implementation of the standard limitted version of Texas Hold’m, sometimes referred to as ‘Limit Texas Hold’em’.

Our implementation wraps RLCard and you can refer to its documentation for additional details. Please cite their work if you use this game in research.

Observation Space

The observation is a dictionary which contains an 'obs' element which is the usual RL observation described below, and an 'action_mask' which holds the legal moves, described in the Legal Actions Mask section.

The main observation space is a vector of 72 boolean integers. The first 52 entries depict the current player’s hand plus any community cards as follows

Legal Actions Mask

The legal moves available to the current agent are found in the action_mask element of the dictionary observation. The action_mask is a binary vector where each index of the vector represents whether the action is legal or not. The action_mask will be all zeros for any agent except the one whos turn it is. Taking an illegal move ends the game with a reward of -1 for the illegally moving agent and a reward of 0 for all other agents.

Action Space

Rewards