Investigating the Impact of Subgraph Social Structure Preference on the Strategic Behavior of Networked Mixed-Motive Learning Agents

2026-04-04

AI Summary

This study investigates how strategic behavior of networked multi-agent systems in sequential social dilemmas is influenced by preferences for specific subgraph-based social structures, such as degree centrality, clustering, and critical links. To this end, the authors propose a Social Relational Intrinsic Motivation (SRIM) mechanism, establishing for the first time a direct link between subgraph structural preferences and multi-agent strategic behavior. They also introduce the Betweenness-Centrality-based Index (BCI) to quantify differences in group-level network structure. Evaluated within reinforcement learning–based multi-agent frameworks in the Harvest and Cleanup environments, the results demonstrate that distinct subgraph preferences significantly affect agents’ payoffs and cooperation levels. Notably, the BCI maintains consistent rankings across environments under identical topologies, indicating that structural preferences exhibit robustness and generality.

Abstract

Limited work has examined the strategic behaviors of relational networked learning agents under social dilemmas, and has overlooked the intricate social dynamics of complex systems. We address the challenge with Socio-Relational Intrinsic Motivation (SRIM), which endows agents with diverse preferences over sub-graphical social structures in order to study the impact of agents' personal preferences over their sub-graphical relations on their strategic decision-making under sequential social dilemmas. Our results in the Harvest and Cleanup environments demonstrate that preferences over different subgraph structures (degree-, clique-, and critical connection-based) lead to distinct variations in agents' reward gathering and strategic behavior: individual aggressiveness in Harvest and individual contribution effort in Cleanup. Moreover, agents with different subgraphical structural positions consistently exhibit similar strategic behavioral shifts. Our proposed BCI metric captures structural variation within the population, and the relative ordering of BCI across social preferences is consistent in Harvest and Cleanup games for the same topology, suggesting the subgraphical structural impact is robust across environments. These results provide a new lens for examining agents' behavior in social dilemmas and insight for designing effective multi-agent ecosystems composed of heterogeneous social agents.

Problem

• social dilemmas
• networked agents
• subgraph structure
• strategic behavior
• social preferences

Innovation

• Socio-Relational Intrinsic Motivation
• subgraph social structure
• multi-agent strategic behavior
• sequential social dilemmas
• BCI metric