ABSTRACT:

The novel coronavirus (COVID-19), which causes an acute respiratory disease in humans, has emerged as a global pandemic, and caused an over 250,000 death toll in the world. Our lab recently launched a project (https://covid19.uclaml.org) to use machine learning to better understand the spread of COVID-19 and further facilitate the decision making of the government agencies.

In this talk, I will focus on an epidemic model-guided machine learning approach for the confirmed case and death forecasts for COVID-19, and peak date projection in both state and national level. In specific, we found that standard epidemic models such as SIR and SEIR are insufficient for modeling the spread of COVID-19.

We therefore propose a variant of the SEIR model that takes into account the untested/unreported cases of COVID-19, and then use a machine learning algorithm to train this model. Validation based on a week ahead prediction indicates that our model is more accurate than many other models including the model proposed by IHME at the University of Washington.

BIO:

Quanquan Gu is an Assistant Professor of Computer Science at UCLA. His current research is in the area of artificial intelligence and machine learning, with a focus on developing and analyzing nonconvex optimization algorithms for machine learning and building the theoretical foundations of deep learning. He received his Ph.D. degree in Computer Science from the University of Illinois at Urbana-Champaign in 2014. He is a recipient of the Yahoo! Academic Career Enhancement Award, NSF CAREER Award, Simons Berkeley Research Fellowship, Adobe Data Science Research Award, Salesforce Deep Learning Research Award and AWS Machine Learning Research Award.

Via Zoom Webinar