The current pandemic provides scientists the opportunity to study epidemiology like never before. While compartmental and agent-based models are reliable tools for simulating epidemics, the current level of technology and connectedness give scientists an immense amount of data to analyze and, in turn, tune these models. In this talk, I will discuss SINDy, a data-driven model discovery tool developed at the University of Washington. The overview of SINDy will be primarily example-driven and will culminate in a real-world scenario as well as those derived from Agent-Based Epidemic simulations. This will lead to a discussion of circumstances in which we might find model discovery useful, even in a field where well-established models exist.
