Knowledge-Based Paranoia Search in Trick-Taking

TL;DR

This paper introduces a knowledge-based paranoia search algorithm for Skat that combines game-tree search with knowledge reasoning, enabling AI to identify forced wins and outperform humans in gameplay.

Contribution

It presents a novel knowledge-based search method for trick-taking games, specifically Skat, integrating partial information reasoning with game-tree search for improved AI performance.

Findings

AI with KBPS outperforms humans in Skat tournaments

Achieves an average score over 1,000 points in standard evaluations

Effective in finding forced wins against most belief states

Abstract

This paper proposes \emph{knowledge-based paraonoia search} (KBPS) to find forced wins during trick-taking in the card game Skat; for some one of the most interesting card games for three players. It combines efficient partial information game-tree search with knowledge representation and reasoning. This worst-case analysis, initiated after a small number of tricks, leads to a prioritized choice of cards. We provide variants of KBPS for the declarer and the opponents, and an approximation to find a forced win against most worlds in the belief space. Replaying thousands of expert games, our evaluation indicates that the AIs with the new algorithms perform better than humans in their play, achieving an average score of over 1,000 points in the agreed standard for evaluating Skat tournaments, the extended Seeger system.