Announcements

• Colloquium on Digital Transformation Science

• 
  

  August 613, 3 pm CT Optimal Targeted Lockdowns in a Multi-Group SIR Model Predictive and Prescriptive Analytics for the COVID-19 Pandemic Dimitris Bertsimas, Boeing Professor of Operations Research and Associate Dean of Business AnalyticsDaron Acemoglu, Institute Professor, Department of Economics, MIT REGISTER FOR ZOOM WEBINARThis colloquium will investigate targeted lockdowns using a multi-group SIR model, in which infection, hospitalization, and fatality rates vary among groups—in particular among young, middle-aged, and old patients. The model—developed by PI Daron Acemoglu with MIT Economics Professors Victor Chernozhukov, Iván Werning, and Michael Whinston—enables a tractable quantitative analysis of optimal policy. For baseline parameter values for the COVID-19 pandemic as applied to the United States, Daron Acemoglu and his colleagues find that optimal policies differentially targeting risk/age groups significantly outperform optimal uniform policies and most of the gains can be realized by having stricter lockdown policies on the oldest group. Intuitively, a strict and long lockdown for the most vulnerable group both reduces infections and enables less strict lockdowns for the lower risk groups. The colloquium also will investigate: the impacts of group distancing, testing, and contract tracing; the matching technology; and the expected arrival time of a vaccine for optimal policies. Overall, Acemoglu’s model indicates targeted policies combined with measures that reduce interactions among groups and increase testing and isolation of the infected can minimize both economic losses and The COVID-19 pandemic creates unprecedented challenges for healthcare providers and policy makers. How to triage patients when healthcare resources are limited? Whom to test? And how to design social distancing policies to contain the disease and its socioeconomic impact? Dimitris Bertsimas and Alexandre Jacquillat of MIT Sloan School of Management believe that analytics can provide an answer and have collected data from clinical studies, case counts, and hospital collaborations at www.covidanalytics.io. This colloquium will present their epidemiological model of the disease’s dynamics, a machine-learning model of mortality risk, and a resource allocation model. It will address: How can we predict admissions in intensive care units using machine learning? How does COVID-19 impact different demographic and socioeconomic populations? How does mobility impact the disease’s spread? How to optimize social distancing policies? How to augment COVID-19 tests with data-driven warnings that identify high-risk subjects? Bertsimas will present a new machine learning model for predicting being COVID-positive (https://covidanalytics.io/infection_calculator) and mortality (https://covidanalytics.io/mortality_calculator) using data from over 40 hospitals around the world, along with high-performance computing (using the C3 AI Suite), and advanced machine learning and artificial intelligence. He will summarize his research group’s end-to-end ML/AI methods, spanning epidemiological modeling (to model the disease’s spread), machine learning (to predict ICU admissions and test results), causal inference (to investigate disparities across populations), and optimal control (to support social distancing guidelines), as well as a new optimization model for allocating vaccines to minimize deaths.

  
  

Quick Links:

C3.ai DTI Webpage

Events

Information on Call for Proposals

Proposal Matchmaking 

Training Materials (password protected)

Have Questions? Please contact one of us:

• Jay Roloff, jayr@illinois.edu (Executive Director, c3.ai.DTI)
• R. Srikant, rsrikant@illinois.edu (Co-Director, c3.ai.DTI)
• Tandy Warnow, warnow@illinois.edu (Co-chief Scientist, c3.ai.DTI)