Learning Individual Behavior in Agent-Based Models with Graph Diffusion Networks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Cozzi, Francesco, Pangallo, Marco, Perotti, Alan, Panisson, André, Monti, Corrado
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911287921868800
author Cozzi, Francesco
Pangallo, Marco
Perotti, Alan
Panisson, André
Monti, Corrado
author_facet Cozzi, Francesco
Pangallo, Marco
Perotti, Alan
Panisson, André
Monti, Corrado
contents Agent-Based Models (ABMs) are powerful tools for studying emergent properties in complex systems. In ABMs, agent behaviors are governed by local interactions and stochastic rules. However, these rules are, in general, non-differentiable, limiting the use of gradient-based methods for optimization, and thus integration with real-world data. We propose a novel framework to learn a differentiable surrogate of any ABM by observing its generated data. Our method combines diffusion models to capture behavioral stochasticity and graph neural networks to model agent interactions. Distinct from prior surrogate approaches, our method introduces a fundamental shift: rather than approximating system-level outputs, it models individual agent behavior directly, preserving the decentralized, bottom-up dynamics that define ABMs. We validate our approach on two ABMs (Schelling's segregation model and a Predator-Prey ecosystem) showing that it replicates individual-level patterns and accurately forecasts emergent dynamics beyond training. Our results demonstrate the potential of combining diffusion models and graph learning for data-driven ABM simulation.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21426
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Individual Behavior in Agent-Based Models with Graph Diffusion Networks
Cozzi, Francesco
Pangallo, Marco
Perotti, Alan
Panisson, André
Monti, Corrado
Artificial Intelligence
Machine Learning
Multiagent Systems
Econometrics
Physics and Society
Agent-Based Models (ABMs) are powerful tools for studying emergent properties in complex systems. In ABMs, agent behaviors are governed by local interactions and stochastic rules. However, these rules are, in general, non-differentiable, limiting the use of gradient-based methods for optimization, and thus integration with real-world data. We propose a novel framework to learn a differentiable surrogate of any ABM by observing its generated data. Our method combines diffusion models to capture behavioral stochasticity and graph neural networks to model agent interactions. Distinct from prior surrogate approaches, our method introduces a fundamental shift: rather than approximating system-level outputs, it models individual agent behavior directly, preserving the decentralized, bottom-up dynamics that define ABMs. We validate our approach on two ABMs (Schelling's segregation model and a Predator-Prey ecosystem) showing that it replicates individual-level patterns and accurately forecasts emergent dynamics beyond training. Our results demonstrate the potential of combining diffusion models and graph learning for data-driven ABM simulation.
title Learning Individual Behavior in Agent-Based Models with Graph Diffusion Networks
topic Artificial Intelligence
Machine Learning
Multiagent Systems
Econometrics
Physics and Society
url https://arxiv.org/abs/2505.21426