Graph Neural Network Surrogates to leverage Mechanistic Expert Knowledge towards Reliable and Immediate Pandemic Response
This provides a faster, reliable tool for time-critical pandemic response, though it is incremental as it applies an existing method to a specific domain.
The authors tackled the bottleneck of slow mechanistic models for pandemic decision-making by developing a graph neural network surrogate that accelerates evaluation by up to 28,670 times, achieving 10-27% mean absolute percentage error across various scenarios.
During the COVID-19 crisis, mechanistic models have guided evidence-based decision making. However, time-critical decisions in a dynamical environment limit the time available to gather supporting evidence. We address this bottleneck by developing a graph neural network (GNN) surrogate of a spatially and demographically resolved mechanistic metapopulation simulator. This combined approach advances classical machine learning approaches which are often black box. Our design of experiments spans outbreak and persistent-threat regimes, up to three contact change points, and age-structured contact matrices on a 400-node spatial graph. We benchmark multiple GNN layers and identify an ARMAConv-based architecture that offers a strong accuracy-runtime trade-off. Across 30-90 day horizons and up to three contact change points, the surrogate attains 10-27 % mean absolute percentage error (MAPE) while delivering (near) constant runtime with respect to the forecast horizon. Our approach accelerates evaluation by up to 28,670 times compared with the mechanistic model, allowing responsive decision support in time-critical scenarios and straightforward web integration. These results show how GNN surrogates can translate complex metapopulation models into immediate, reliable tools for pandemic response.