Yuqi Gu joined the Department of Statistics at Columbia University as a tenure-track Assistant Professor in July 2021. Prior to this, she spent a year from 2020 to 2021 as a postdoctoral researcher at Duke University, under the supervision of David B. Dunson. In 2020 she received a Ph.D. in Statistics from the University of Michigan, advised by Gongjun Xu. In 2015 she received a B.S. in Mathematics from Tsinghua University.
Prof. Gu is broadly interested in latent variable models, including their various variants in statistical machine learning and their applications in social and biomedical sciences. Her current research interests include methods and theory of complicated latent structures, mixed membership models, multivariate categorical data, psychometrics and cognitive diagnostic modeling; and she is also very interested in the related fields of tensor decompositions, graphical models, mixture models, and Bayesian statistics. In these interrelated areas, Prof. Gu enjoys building latent structure models with interpretability and identifiability, studying the intriguing theoretical properties induced by marginalizing out the latent variables, and developing efficient and rigorous methods applicable to large-scale and high-dimensional real data.