AAMAS 2023 Imperfect-Information Card Games Competition

-->

Background

Mahjong, Texas hold’em and Bridge are three popular strategical board games worldwide. Each game has plenty of advanced players, let alone the amateurs. These games often have complex game rules with enormously large state space, rendering it challenging for designing a strong AI player. There have been various attempts to solve these classic card games. In Mahjong, Suphx developed by Microsoft Research Asia is the first AI system that outperforms most top human players using deep reinforcement learning methods; in the Heads-Up No-Limit Texas Hold’em game, AlphaHoldem manages to reach the level of professional human players through self-playing; in the multi-player Texas Hold’em game, Pluribs is trained to play a strong strategy that is capable of consistently defeating elite human professionals; in the game of Bridge, there exists a well-known computer bridge software Wbridge5 using a Monte-Carlo search in the bidding process and a few attempts adopting deep reinforcement learning techniques.

In order to further advance the research in the turn-based card game AI, we propose the AAMAS 2023 Imperfect-Information Card Games Competition and hold the competition on the Jidi platform. This competition aims to encourage the development of card game AI under imperfect-information setting when facing various opponent tactics.

Tracks

Track - 1: Four-player No-limit Texas Hold'em Competition

The Four-player Texas Hold’em is a betting game. Each of the four players is initially dealt two face-down cards (called the hole cards). Then the dealer deals sequentially five face-up community cards in three stages: the flop, the turn and the river. Each player pursues the best five-card combination among the hole cards and the community cards. Each player can choose to call, to check, to raise or to fold during each round. The participant needs to design a betting agent that can win as many chips as possible after rounds of play.

Track - 2: Bridge Competition

Bridge is a popular four-person card game using a regular 52 card deck with ace being high. This is a 2v2 game where the north and the south players are partners against the partnership of the east and the west players. A game round consists of a bidding round and a card playing round. The participant needs to design a single agent that can cooperate with an assigned teammate and win as many game rounds as possible.

Track - 3: Mahjong Competition

Mahjong is a four-player tile-based game. It uses a deck of 136 tiles consists of 4 identical sets of 34 unique tiles. The goal is to complete a legal hand (4 sets and a pair) within 14 draws. Each player can choose to chow, pong, kong or discard a particular tile. The participant needs to design an agent that can win as many games round as possible.

Prize for each track

Champion

£ TBD

2nd

£ TBD

3rd

£ TBD

Participation

Come and Sign up for the competition!

We provide Jidi tutorial and source code for the competition

Timeline

Open Registration: 2023.04.01, 10:30 (UTC+8)
Warm-up 1st Round (Submission Deadline 2023.04.10 10:30 UTC+8): Swiss Round double competition system, the result is not included in the final competition result
Warm-up 2st Round (Submission Deadline 2023.04.17 10:30 UTC+8): Swiss Round double competition system, the result is not included in the final competition result
Final 1st Round (Submission Deadline 2023.05.01 10:30 UTC+8): Swiss Round double competition system, score multiplied by 0.3 is included in the final total score
Final 2st Round (Submission Deadline 2023.05.15 10:30 UTC+8): Swiss Round double competition system, score multiplied by 0.7 is included in the final total score

Organizers

Haifeng Zhang

Associate Professor at Institute of Automation, Chinese Academy of Sciences

Nanjing Artificial Intelligence Research of IA

Email: haifeng.zhang@ia.ac.cn

Yan Song

Software Engineer at Institute of Automation, Chinese Academy of Sciences

Email: yan.song@ia.ac.cn

Kun Shao

Principal Research Scientist at Huawei Noah’s Ark Lab

Email: shaokun2@huawei.com

David Mguni

Principal Research Scientist at Huawei Noah’s Ark Lab

Email: david.mguni@huawei.com


br>