Computing Games: Bridging the Gap between Search and Entertainment

Abstract

Over history, games have served multiple purposes. It serves as a fun activity for players who need the entertainment to become test-beds for artificial intelligence. Solving games is beneficial in providing a better understanding of how information is progressing throughout the game. Uncertainty in games affects the way a game is solved and the way the game is experienced. Previous works have interpreted uncertainty in the game progress through various means, but there have been no clear links among those interpretations. In this paper, the probability-based proof number search (PPNS) and single conspiracy number (SCN) were used as the domain-independent indicators to analyze how uncertainty affects various game elements. PPNS exploits information from certain and uncertain information to reach convergence in solving games. Meanwhile, SCN evaluates the game states' difficulty and describes game-playing patterns to understand play positions better. The study's objective focuses on finding the optimal difficulty ordering of a game solver, defining the indicator for entertainment, and linking game-tree search and entertainment in different game environments. Experiments results demonstrate the link between the search indicators and the measure of entertainment where uncertainty plays a vital role in both contexts, verified from both two-person and single-agent games. Such a situation is also crucial for both computation and entertainment measures since it impacts both the quality of information and the expected game-playing experience.