Presentation on YouTube

COVID-19 pandemic showed that while designing policies to mitigate an epidemic, regulators should consider the response of large number of interacting agents to these policies. To address this, we propose a Stackelberg mean field game (MFG) model between a government who optimizes mitigation policies and many non-cooperative individuals who chooses their socialization levels to control their jumps between different health states. Nash equilibrium response to the policies is approximated with an MFG. We will give equilibrium characterization and introduce a numerical approach based on machine learning tools. Finally, we will discuss an extension where an underlying network among individuals is implemented via graphon game.