| Abstract: |
| A complex system refers to a collection of interconnected elements or components that exhibit emergent behaviors and properties arising from their interactions. The COVID-19 pandemic, as an example of such a complex system, highlighted the significance of mathematical epidemiological modeling and data analysis in understanding disease dynamics. It also highlighted the importance of interdisciplinary collaborations integrating research on behavioral and/or social processes in mathematical epidemiological models with a goal to minimize unintended outcomes of public health interventions in response to pandemics. In this work we aim to enhance complex epidemiological models by integrating insights from social and behavioral sciences with data-driven predictive analytics. The proposed work will present a hybrid framework combining Agent-Based Modeling, Compartmental Models, and Physics-Informed Neural Networks applied to benchmark problems to efficiently conduct parameter estimation and data-driven decision-making. |
|