Multi-Agent Digital Twins for Strategic Decision-Making using Active Inference

TL;DR

This paper extends active inference to multi-agent digital twins, enabling adaptive, goal-oriented decision-making in complex, dynamic environments with decentralized models.

Contribution

It introduces a novel multi-agent framework with contextual inference and streaming machine learning, enhancing adaptability and scalability in digital twin applications.

Findings

Demonstrated the framework with a Cournot competition example.

Showed improved adaptability in dynamic environments.

Enabled coordinated decision-making among agents.

Abstract

Active Inference is an emerging framework providing a quantitative account of behavioral processes in neuroscience and a principled approach to decision-making under uncertainty. Its application to agency problems is natural, offering an autopoietic interpretation of action while addressing classical challenges such as the exploration-exploitation trade-off. Recently, Active Inference has been applied to digital twin scenarios for adaptive and predictive modeling of complex systems. In this work, we extend Active Inference to multi-agent digital twins in which agents interact within a shared environment while maintaining decentralized generative models. Our multi-agent framework features two innovations: (i) contextual inference to improve adaptability in dynamic environments, and (ii) the integration of streaming machine learning within agents' generative structures, enabling tunable goal-oriented behavior while preserving efficiency and scalability. The framework is illustrated through a Cournot competition example, providing a digital twin representation of a socio-economic system and highlighting its potential for coordinated decision-making in multi-agent contexts.