Towards agent-based-model informed neural networks

2025-12-05

AI Summary

Standard neural differential equations (Neural ODEs) struggle to incorporate structural constraints—such as mass conservation, locality, or bounded rationality—in domains lacking explicit physical conservation laws (e.g., energy). To address this, we propose ABM-NN: an agent-based modeling–inspired, differentiable neural architecture that embeds structural priors. ABM-NN integrates constrained graph neural networks, hierarchical decomposition, and neural differential equations to explicitly enforce mass conservation and local interaction constraints, enabling counterfactual reasoning and intervention analysis. Experiments on the Lotka–Volterra system, graph-based SIR epidemic dynamics, and macroeconomic time series from the ten largest economies demonstrate that ABM-NN significantly outperforms baselines—including GCN and GraphSAGE—achieving superior robustness under strong noise, enhanced extrapolation capability, and improved physical consistency.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

In this article, we present a framework for designing neural networks that remain consistent with the underlying principles of agent-based models. We begin by highlighting the limitations of standard neural differential equations in modeling complex systems, where physical invariants (like energy) are often absent but other constraints (like mass conservation, network locality, bounded rationality) must be enforced. To address this, we introduce Agent-Based-Model informed Neural Networks(ABM-NNs), which leverage restricted graph neural networks and hierarchical decomposition to learn interpretable, structure-preserving dynamics. We validate the framework across three case studies of increasing complexity: (i) a generalized Generalized Lotka--Volterra system, where we recover ground-truth parameters from short trajectories in presence of interventions; (ii) a graph-based SIR contagion model, where our method outperforms state-of-the-art graph learning baselines (GCN, GraphSAGE, Graph Transformer) in out-of-sample forecasting and noise robustness; and (iii) a real-world macroeconomic model of the ten largest economies, where we learn coupled GDP dynamics from empirical data and demonstrate gradient-based counterfactual analysis for policy interventions.

Problem

• Design neural networks consistent with agent-based model principles
• Enforce constraints like mass conservation and bounded rationality
• Learn interpretable dynamics for complex systems via ABM-NNs

Innovation

• Agent-based-model informed neural networks enforce constraints
• Restricted graph neural networks learn interpretable dynamics
• Hierarchical decomposition preserves structure for complex systems