16702

Video URL

https://simons.berkeley.edu/talks/country-scale-bandit-implementation-targeted-covid-19-testing

Country-Scale Bandit Implementation for Targeted COVID-19 Testing

(2020). Country-Scale Bandit Implementation for Targeted COVID-19 Testing. The Simons Institute for the Theory of Computing. https://simons.berkeley.edu/talks/country-scale-bandit-implementation-targeted-covid-19-testing

Country-Scale Bandit Implementation for Targeted COVID-19 Testing. The Simons Institute for the Theory of Computing, Oct. 29, 2020, https://simons.berkeley.edu/talks/country-scale-bandit-implementation-targeted-covid-19-testing 

          @misc{ scivideos_16702,
            doi = {},
            url = {https://simons.berkeley.edu/talks/country-scale-bandit-implementation-targeted-covid-19-testing},
            author = {},
            keywords = {},
            language = {en},
            title = {Country-Scale Bandit Implementation for Targeted COVID-19 Testing},
            publisher = {The Simons Institute for the Theory of Computing},
            year = {2020},
            month = {oct},
            note = {16702 see, \url{https://scivideos.org/Simons-Institute/16702}}
          }
Hamsa Bastani (Wharton School of the University of Pennsylvania)
Talk number16702
Source RepositorySimons Institute
Subject

Abstract

In collaboration with the Greek government, we use machine learning to manage the threat of COVID-19. With tens of thousands of international visitors every day, Greece cannot test each visitor to ensure that they are not a carrier of COVID-19. We developed a bandit policy that balances allocating scarce tests to (i) continuously monitor the dynamic infection risk of passengers from different locations (exploration), and (ii) preferentially target risky tourist profiles for testing (exploitation). Our solution is currently deployed across all ports of entry to Greece. I will describe a number of technical challenges, including severely imbalanced outcomes, batched/delayed feedback, high-dimensional arms, port-specific testing constraints, and transferring knowledge from (unreliable) public epidemiological data. Joint work with Kimon Drakopoulos and Vishal Gupta.