A Computable Game-Theoretic Framework for Multi-Agent Theory of Mind

TL;DR

This paper introduces a game-theoretic computational framework for multi-agent Theory of Mind, enabling bounded rational decision-making with recursive beliefs while ensuring computational feasibility through statistical and approximation methods.

Contribution

It formalizes a game-theoretic approach to multi-agent ToM, integrating bounded rationality and recursive belief modeling with computationally practical techniques.

Findings

Framework allows recursive beliefs about others' mental states.

Employs statistical and approximation methods for computability.

Supports decision-making in multi-agent systems with ToM.

Abstract

Originating in psychology, $\textit{Theory of Mind}$ (ToM) has attracted significant attention across multiple research communities, especially logic, economics, and robotics. Most psychological work does not aim at formalizing those central concepts, namely $\textit{goals}$, $\textit{intentions}$, and $\textit{beliefs}$, to automate a ToM-based computational process, which, by contrast, has been extensively studied by logicians. In this paper, we offer a different perspective by proposing a computational framework viewed through the lens of game theory. On the one hand, the framework prescribes how to make boudedly rational decisions while maintaining a theory of mind about others (and recursively, each of the others holding a theory of mind about the rest); on the other hand, it employs statistical techniques and approximate solutions to retain computability of the inherent computational problem.