Challenging Human Supremacy: Evaluating Monte Carlo Tree Search and Deep Learning for the Trick Taking Card Game Jass

Despite the recent successful application of Artificial Intelligence (AI) to games, the performance of cooperative agents in imperfect information games is still far from surpassing humans. Cooperating with teammates whose play-styles are not previously known poses additional challenges to current state-of-the-art algorithms. In the Swiss card game Jass, coordination within the two opposing teams is crucial for winning. Since verbal communication is forbidden, the only way to transmit information within the team is through a player’s play-style. This makes the game a particularly suitable candidate subject to continue the research on AI in cooperation games with hidden information. In this work, we analyse the effectiveness and shortcomings of several state-of-the-art algorithms (Monte Carlo Tree Search (MCTS) variants and Deep Neural Networks (DNNs)) at playing the Jass game. Our key contributions are two-fold. First, we provide a performance overview for state-of-the-art algorithms, thus, setting a strong foundation for further research on the subject. Second, we implement and open-source1 a platform where different agents (both humans and AI) can play Jass in an automated fashion, effectively reducing the overhead for other researchers who want to perform further experiments.