2021/01/04 by Lawrence Thul, Thul, Lawrence, Warren B. Powell +1
Decision Sciences · Mathematics · Medicine · #Advanced Bandit Algorithms Research #Artificial Intelligence (cs.AI) #COVID-19 epidemiological studies #FOS: Computer and information sciences #Machine Learning (cs.LG) #SARS-CoV-2 and COVID-19 Research
paper · pdf · doi:10.48550/arxiv.2101.01204
openalex publication_date 2021/01/04 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
The pandemic caused by the SARS-CoV-2 virus has exposed many flaws in the\ndecision-making strategies used to distribute resources to combat global health\ncrises. In this paper, we leverage reinforcement learning and optimization to\nimprove upon the allocation strategies for various resources. In particular, we\nconsider a problem where a central controller must decide where to send testing\nkits to learn about the uncertain states of the world (active learning); then,\nuse the new information to construct beliefs about the states and decide where\nto allocate resources. We propose a general model coupled with a tunable\nlookahead policy for making vaccine allocation decisions without perfect\nknowledge about the state of the world. The lookahead policy is compared to a\npopulation-based myopic policy which is more likely to be similar to the\npresent strategies in practice. Each vaccine allocation policy works in\nconjunction with a testing kit allocation policy to perform active learning.\nOur simulation results demonstrate that an optimization-based lookahead\ndecision making strategy will outperform the presented myopic policy.\n