BEGIN:VCALENDAR BEGIN:VEVENT SUMMARY:Infectious Disease Forecasting: Methods, Lessons, and Opportunities DESCRIPTION:Speaker: Dr. Ajitesh Srivastava, Research Assistant Professor/Ming Hsieh Department of Computer and Electrical Engineering Talk Title: Infectious Disease Forecasting: Methods, Lessons, and Opportunities Abstract: After more than a year, COVID-19 continues to disrupt our lives. Efforts have been underway since the beginning to understand the epidemiological situation and generate short-term forecasts and long-term scenario projections to drive public health decisions. These efforts, called "hubs" are collaborations between government agencies and multiple universities. In this talk, I will discuss the lessons learned from my participation in several such efforts in modeling and projection of COVID-19, and the resulting research opportunities. I will also present my methodology which has evolved over time to now incorporate dynamics of multiple competing variants, vaccination behavior, age-specific contact matrices, and waning immunity. A key feature of the approach is that it avoids overfitting by splitting the model into independent linear regression problems. An additional advantage is that the runtime is low. As an example, learning the model and generating case, death, and hospitalization forecasts for 56 regions of the US, each with around 25 variants and 5 age groups, takes ~20s on a 2-core desktop. This also enables fast scenario projections, where even for each scenario multiple runs are needed to incorporate uncertainty in hyper-parameters and human behavior. This may not be the last pandemic we will face, and therefore the research does not end with COVID-19. In fact, the extensive data-collection, monitoring, and forecasting during this epidemic sets the stage for more impactful research in preparedness for future epidemics. Biography: Dr. Ajitesh Srivastava is a Research Assistant Professor at the Ming Hsieh Department of Computer and Electrical Engineering. He obtained his PhD in Computer Science at the University of Southern California in 2018. His research interests include network science, modeling, and machine learning applied to epidemics, social good, social networks, and systems. He collaborates with teams around the world, the CDC, and the ECDC for infectious disease forecasting and scenario projections. He is a DARPA Grand Challenge Winner (2014) on predicting the spread of Chikungunya virus. Host: Dr. Richard M. Leahy, Chair, Ming Hsieh Department of Electrical and Computer Engineering (Systems) Webcast: https://usc.zoom.us/j/91552972911?pwd=VG5DczVLdk9vQllBK2ZQT2l3dUJuQT09 DTSTART:20211102T100000 LOCATION:EEB 248 URL;VALUE=URI: DTEND:20211102T110000 END:VEVENT END:VCALENDAR