This report from Global Partnership on Artificial Intelligence (GPAI)'s Pandemic Resilience project follows its 2023 report and is focused on practically implementing the concepts previously developed by the project team. Indeed, the 2023 report laid the foundation for this research while presenting recommendations on various approaches that aligned pandemic modelling with responsible Artificial Intelligence (AI). The 2023 report showcased a calibration framework approach and an ensemble modelling concept, focusing on the added value and pertinence of both consistent calibration and ensembling; that is, ensuring models are consistent in shared parameter values while using the strengths of different models and creating a digital "task force". The combination of the calibration framework and ensemble model encourages and enables modellers from different locations and backgrounds to work together by using standardised versions of their work. Although there has been substantial modelling activity of Non-Pharmaceutical Interventions (NPIs) for COVID-19, this activity has been fragmented across different countries, with mixed access and sharing of data and models. This report documents a prototype calibration framework - based on a multi-objective genetic algorithm - that simultaneously calibrates multiple models across different locations and ensures consistent parameter values across models. The resulting, calibrated models are then combined using an ensemble modelling concept that provides more accurate model results than any of the models do individually. Hence, consistent models for multiple locations are created and can be shared easily with these locations. In addition, diverse perspectives from the models can provide more accurate results for each location through the ensemble model. Initial case study results show that long runs of the calibration framework improves model accuracy by approximately 60%. They also show the efficacy of the calibration framework over manual calibration. However, artefacts from the underlying models still present challenges for calibration at the beginning of a modelling horizon. Initial case study results for the ensemble model show reasonable improvements for prediction accuracy in locations with large numbers of COVID-19 cases, but the inclusion of models that work well for lower case numbers needs to be explored to fully investigate the benefits and limitations of ensemble modelling.

PDF

Technical Report, Global Partnership on Artificial Intelligence, November 2024.

@techreport{gpai:report24, title={Pandemic Resilience: Case studies of an AI-calibrated ensemble of models to inform decision making}, author={GPAI}, month={November}, institution={Global Partnership on Artificial Intelligence}, type={Report}, url="http://nn.cs.utexas.edu/?gpai:report24", year={2024} }