Selecting Robust Strategies When Players Do Not Know Exactly What Game They Are Playing
Abstract
Game theory is a tool for modeling multi-agent decision problems and has been used to great success in modeling and simulating problems such as poker, security, and trading agents. However, many real games are extremely large and complex with multiple agent interactions. One approach for solving these games is to use abstraction techniques to shrink the game to a form that can be solved by removing details and translating a solution back to the original. However, abstraction introduces error into the model. This research studies ways to analyze games, abstractions, and strategies that are robust to noise in the game.Gaining a better understanding of the effect of abstraction in game-theoretic analysis requires a focus on the strategy selection problem: how should an agent choose a strategy to play in a game, based on an abstracted game model? This problem has three interacting components: (1) the method for game abstraction, (2) the strategy solution method, and (3) the method for reverse mapping this strategy. This approach has been studied extensively for poker, which is a 2-player, zero-sum game. However, much less is known about how abstraction interacts with strategy selection in more general games.This research introduces a formal model for studying the situation where players select strategies based on noisy game models, including abstraction. I use two generalized types of abstraction techniques and over a dozen found from the literature. Experimental results of tournaments between agents using abstraction on different classes of games demonstrates that each of these elements has a strong influence on the results. Then I design and develop new agents that are robust against the noise in games.This research has identified properties of robust solution techniques, evaluated the impact of abstraction on solution quality, presented useful definitions of robustness, and created a new family of solution techniques that allow players to select strategies that are more robust to noise in game models.
Subject Area
Artificial intelligence|Computer science|Information science
Recommended Citation
Veliz, Oscar Samuel, "Selecting Robust Strategies When Players Do Not Know Exactly What Game They Are Playing" (2021). ETD Collection for University of Texas, El Paso. AAI28718062.
https://scholarworks.utep.edu/dissertations/AAI28718062