Variational Inference with Agent-Based Models

TL;DR

This paper introduces a variational inference method for agent-based models that enables tracking and predicting complex real-world systems from imperfect data, exemplified by epidemic spread and traffic congestion.

Contribution

It presents a novel variational approach integrating big data with agent-based models for improved system tracking and prediction.

Findings

Effective epidemic tracking at the individual level

Accurate short-term traffic congestion predictions

Bridges modelers and data miners in a unified framework

Abstract

In this paper, we develop a variational method to track and make predictions about a real-world system from continuous imperfect observations about this system, using an agent-based model that describes the system dynamics. By combining the power of big data with the power of model-thinking in the stochastic process framework, we can make many valuable predictions. We show how to track the spread of an epidemic at the individual level and how to make short-term predictions about traffic congestion. This method points to a new way to bring together modelers and data miners by turning the real world into a living lab.