This project explores the role of Theory of Mind (ToM) in in collaborative contexts, with a particular focus on its application in the Hanabi card game. We investigate how transparent decision-making can improve agent coordination by modeling unobservable mental states such as desires, beliefs, and intentions.
We compare two methodologies for modeling ToM: a logic-based approach and a decision-tree-based approach. Both approaches integrate beliefs about players' cards, generated through the Hand Card Information Completion module, into the decision-making process.
The logic-based approach uses logical inference to deduce optimal actions from past events and contextual clues. The decision-tree-based approach breaks down decision-making into hierarchical levels, improving efficiency in navigating the decision space.
The goal of this project is to evaluate the effectiveness and transparency of these approaches in promoting competent interactions within collaborative environments. This repository contains the code and resources to implement both methods and compare their performance in Hanabi.