Theory of Mind Using Active Inference: A Framework for Multi-Agent Cooperation

2025-08-01

AI Summary

In multi-agent collaboration, modeling others’ beliefs and goals remains challenging without explicit communication or shared models. Method: This paper proposes a Theory of Mind (ToM) framework grounded in active inference, eschewing task-specific generative models and explicit communication. It jointly infers others’ beliefs, goals, and the agent’s own policy via recursive Bayesian inference over observed behaviors, and innovatively extends the inference-tree planning algorithm to enable decoupled internal representations of self- and other-beliefs. Results: Evaluated on obstacle avoidance and foraging tasks, ToM-enabled agents reduce collision rates and behavioral redundancy significantly; collaborative efficiency improves by 27%–41% over ToM-free baselines. The framework demonstrates effectiveness, scalability, and decentralization—offering a general-purpose, communication-free ToM mechanism for autonomous agents.

Technology Category

• Multiagent Systems: Modeling other Agents
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

We present a novel approach to multi-agent cooperation by implementing theory of mind (ToM) within active inference. ToM - the ability to understand that others can have differing knowledge and goals - enables agents to reason about others' beliefs while planning their own actions. Unlike previous active inference approaches to multi-agent cooperation, our method neither relies on task-specific shared generative models nor requires explicit communication, while being generalisable. In our framework, the ToM-equipped agent maintains distinct representations of its own and others' beliefs and goals. We extend the sophisticated inference tree-based planning algorithm to systematically explore joint policy spaces through recursive reasoning. Our approach is evaluated through collision avoidance and foraging task simulations. Results demonstrate that ToM-equipped agents cooperate better compared to non-ToM counterparts by being able to avoid collisions and reduce redundant efforts. Crucially, ToM agents accomplish this by inferring others' beliefs solely from observable behaviour. This work advances practical applications in artificial intelligence while providing computational insights into ToM.

Problem

• Implementing theory of mind in multi-agent cooperation using active inference
• Enabling agents to infer others' beliefs without explicit communication
• Improving cooperation by avoiding collisions and reducing redundant efforts

Innovation

• Active inference with theory of mind
• Recursive reasoning in joint policies
• Belief inference from observable behavior