Calibrating Agent-based Models to Microdata with Graph Neural Networks

TL;DR

This paper introduces a novel method for calibrating agent-based models by directly learning parameter posteriors from granular microdata using temporal graph neural networks, avoiding information loss from summary statistics.

Contribution

It proposes using temporal graph neural networks to directly infer parameters from detailed agent microstates, improving calibration accuracy over traditional summary statistic methods.

Findings

Effective calibration using raw microdata with GNNs

Better preservation of information compared to summary statistics

Demonstrated advantages in Bayesian inference for ABMs

Abstract

Calibrating agent-based models (ABMs) to data is among the most fundamental requirements to ensure the model fulfils its desired purpose. In recent years, simulation-based inference methods have emerged as powerful tools for performing this task when the model likelihood function is intractable, as is often the case for ABMs. In some real-world use cases of ABMs, both the observed data and the ABM output consist of the agents' states and their interactions over time. In such cases, there is a tension between the desire to make full use of the rich information content of such granular data on the one hand, and the need to reduce the dimensionality of the data to prevent difficulties associated with high-dimensional learning tasks on the other. A possible resolution is to construct lower-dimensional time-series through the use of summary statistics describing the macrostate of the system at each time point. However, a poor choice of summary statistics can result in an unacceptable loss of information from the original dataset, dramatically reducing the quality of the resulting calibration. In this work, we instead propose to learn parameter posteriors associated with granular microdata directly using temporal graph neural networks. We will demonstrate that such an approach offers highly compelling inductive biases for Bayesian inference using the raw ABM microstates as output.