James G. Scott

James G. Scott joined The University of Texas at Austin in 2009. He has made foundational contributions to Bayesian methods for high-dimensional data, particularly in the areas of sparsity, regularization, multiple testing, and Bayesian computation.  His recent work has focused on pressing public health and policy questions, including pioneering studies on access to abortion medication via telemedicine, as well as the creation of real-time disease tracking and forecasting models during the COVID-19 pandemic. His current work concerns the statistical reliability of modern generative ML models for scientific inference, with a focus on methods neural posterior estimation. He received the UT System Regents’ Outstanding Teaching Award in 2014, among many other teaching awards.