AILGMar 30, 2025

An Organizationally-Oriented Approach to Enhancing Explainability and Control in Multi-Agent Reinforcement Learning

arXiv:2503.23615v11 citationsh-index: 16AAMAS
Originality Incremental advance
AI Analysis

This work addresses the challenge of enhancing explainability and control at the organizational level in multi-agent systems, which is incremental as it builds on existing organizational concepts.

The paper tackles the problem of limited explainability and control in multi-agent reinforcement learning by introducing a framework that incorporates organizational roles and goals from the MOISE+ model, resulting in demonstrated coherence between predefined and inferred organizational specifications across various environments and algorithms.

Multi-Agent Reinforcement Learning can lead to the development of collaborative agent behaviors that show similarities with organizational concepts. Pushing forward this perspective, we introduce a novel framework that explicitly incorporates organizational roles and goals from the $\mathcal{M}OISE^+$ model into the MARL process, guiding agents to satisfy corresponding organizational constraints. By structuring training with roles and goals, we aim to enhance both the explainability and control of agent behaviors at the organizational level, whereas much of the literature primarily focuses on individual agents. Additionally, our framework includes a post-training analysis method to infer implicit roles and goals, offering insights into emergent agent behaviors. This framework has been applied across various MARL environments and algorithms, demonstrating coherence between predefined organizational specifications and those inferred from trained agents.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes