AIMASIJul 12, 2022

Reward-Sharing Relational Networks in Multi-Agent Reinforcement Learning as a Framework for Emergent Behavior

arXiv:2207.05886v214 citationsh-index: 6
Originality Incremental advance
AI Analysis

This work addresses the challenge of emergent behavior in multi-agent systems for AI researchers, but it is incremental as it builds on existing MARL with a novel relational framework.

The authors tackled the problem of integrating social interactions into multi-agent reinforcement learning by proposing Reward-Sharing Relational Networks (RSRN) to model agent relationships, and found that different network structures significantly influence learned behaviors in a 3-agent scenario.

In this work, we integrate `social' interactions into the MARL setup through a user-defined relational network and examine the effects of agent-agent relations on the rise of emergent behaviors. Leveraging insights from sociology and neuroscience, our proposed framework models agent relationships using the notion of Reward-Sharing Relational Networks (RSRN), where network edge weights act as a measure of how much one agent is invested in the success of (or `cares about') another. We construct relational rewards as a function of the RSRN interaction weights to collectively train the multi-agent system via a multi-agent reinforcement learning algorithm. The performance of the system is tested for a 3-agent scenario with different relational network structures (e.g., self-interested, communitarian, and authoritarian networks). Our results indicate that reward-sharing relational networks can significantly influence learned behaviors. We posit that RSRN can act as a framework where different relational networks produce distinct emergent behaviors, often analogous to the intuited sociological understanding of such networks.

Foundations

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

Your Notes