David B. Dunson

Department of Statistical Science, Mathematics, and ECE, Duke University


David Dusnon.jpg

David Dunson is an Arts & Sciences Distinguished Professor at Duke University.  His research focuses on developing novel data analysis methods motivated by scientific applications, with a particular emphasis on complex and high-dimensional biomedical data.  He has worked broadly on probabilistic and Bayesian methods for multivariate, dependent and “object” data.  A focus is on latent structure learning approaches, methods for including prior information (which may take complex forms) and nonparametric Bayes methods accounting for uncertainty in modeling assumptions.  He has also worked broadly on designing new classes of computationally efficient algorithms.  An active interest is in geometric data science, developing new representations for complex data that can be used in data analysis, modeling and inferences.  He work is highly cited, and he has received numerous awards including the COPSS Presidents’ Award for the top statistician internationally under the age of 41.