Dr. Geoff Pleiss

Biography

Geoff Pleiss is an Assistant Professor at the University of British Columbia in the Department of Statistics and a CIFAR AI Chair at the Vector Institute. He holds a Ph.D. from the Computer Science Department at Cornell University, where he was advised by Kilian Weinberger and worked closely with Andrew Gordon Wilson. His research interests lie at the intersection of deep learning and probabilistic modeling, focusing on heuristic and approximate notions of uncertainty in machine learning models. His major work includes quantifying uncertainty in neural networks, Bayesian optimization, Gaussian processes, and ensemble methods. He is an active contributor to open-source projects and notably co-created and maintains GPyTorch, a Gaussian process library. Before his current role, he completed a postdoctoral fellowship at Columbia University under John P. Cunningham. Geoff aims to inform reliable decision-making processes through experimental design and scientific discovery.

Research Interests