MRC BSU Seminar Series

Speaker: Jesse Clifton (Department of Statistics, North Carolina State University).

Title: Reinforcement learning for large-scale spatio-temporal decisions

Abstract: Emerging infectious diseases are a cause of humanitarian and economic crises around the world. The costs of epidemics such as the 2013-2015 Ebola virus outbreak in West Africa provide strong motivation for the examination of adaptive treatment strategies that allocate resources in response to and anticipation of the evolution of an epidemic. We formalize adaptive management of an emerging infectious disease spreading across a set of locations as a treatment regime that maps up-to-date information on the epidemic to a subset of locations identified as high-priority for treatment. The resulting sequential decision problem poses interesting computational and statistical challenges. These include the need to make use of scientifically-informed dynamics models while guarding against model misspecification; pooling information across spatial locations in an efficient manner; and optimizing over a combinatorially-large space of resource allocations. We present parametric, semiparametric, and hybrid approaches to estimating an optimal regime, and study these in toy simulations as well as simulations based on the 2013-2015 Ebola outbreak.