Learning to Mitigate Epidemic Risks: A Dynamic Population Game Approach

Area of Science:

• Epidemiology
• Game Theory
• Mathematical Modeling

Background:

• Infectious diseases necessitate understanding population behavior and intervention effectiveness.
• Dynamic population games offer a framework to model individual choices during epidemics.

Purpose of the Study:

• To develop a dynamic population game model for epidemic scenarios.
• To analyze individual decision-making regarding vaccination, testing, and social activity.
• To evaluate the impact of public health interventions on disease dynamics.

Main Methods:

• Developed a dynamic population game model with five infection states (susceptible, asymptomatic, symptomatic, recovered, unknowingly recovered).
• Defined agent best response to maximize long-run discounted expected reward.
• Analyzed stationary Nash equilibrium and transient evolution using evolutionary learning dynamics.

Main Results:

• Demonstrated the existence of a stationary Nash equilibrium in the population game.
• Showcased how individuals weigh vaccination, testing, and social activity trade-offs.
• Illustrated the effects of intervention strategies, like social distancing, on vaccination and infection rates.

Conclusions:

• The model provides insights into individual risk-benefit analysis during epidemics.
• Intervention strategies significantly influence vaccination uptake and infection prevalence.
• Dynamic population games are valuable for understanding and managing infectious disease outbreaks.