Accurate Calibration of Agent-based Epidemiological Models with Neural Network Surrogates

TL;DR

This paper introduces a neural network surrogate model for calibrating agent-based epidemiological models across multiple regions, improving accuracy and enabling joint regional parameter estimation.

Contribution

It presents a novel neural network surrogate that emulates multiple locations simultaneously and a new posterior estimation method for better calibration.

Findings

Neural network surrogate accurately emulates epidemiological models across regions.

Joint regional parameter estimation improves calibration accuracy.

Enhanced posterior estimates lead to better disease dynamics understanding.

Abstract

Calibrating complex epidemiological models to observed data is a crucial step to provide both insights into the current disease dynamics, i.e.\ by estimating a reproductive number, as well as to provide reliable forecasts and scenario explorations. Here we present a new approach to calibrate an agent-based model -- EpiCast -- using a large set of simulation ensembles for different major metropolitan areas of the United States. In particular, we propose: a new neural network based surrogate model able to simultaneously emulate all different locations; and a novel posterior estimation that provides not only more accurate posterior estimates of all parameters but enables the joint fitting of global parameters across regions.