Home

Trevor Harris

Assistant Professor
Department of Statistics
University of Connecticut

Statistics, Machine Learning, Uncertainty, and Climate.

Research Areas

Climate Model Evaluation and Integration
My research develops statistical and machine learning methods for evaluating climate models against observations, with a focus on comparing long-run climate distributions and constraining future projections. As a counterpart to this work, I also work on deep learning / neural operator approaches that integrate physical model output with observational data to constrain climate projections and reduce bias.

Uncertainty Quantification and Generative Modeling
I am also working on conformal and distribution-free uncertainty quantification methods for neural operators more generally, with recent efforts connecting conformal inference to generative models that represent full predictive distributions. Ongoing research explores calibrated flow-based generative models as efficient alternatives to autoregressive climate emulators.

Structure Discovery in Climate Systems
Finally, my group is also develops statistical and machine learning methods to uncover dependence structures, causal relationships, and compound risk in large-scale climate systems, enabling more reliable characterization of extremes and rare events.

Bio

I am is an Assistant Professor of Statistics at the University of Connecticut. My research develops statistical methods for learning and inference in complex spatiotemporal systems, with a focus on quantifying and representing uncertainty in machine learning models. Recent work includes probabilistic modeling, calibrated prediction, and operator regression, and I am particularly interested in methods that are reliable when data are high-dimensional, dependent, or imperfectly aligned. This work is motivated by applications in scientific and environmental data analysis, including climate modeling. Prior to joining the University of Connecticut, I was an Assistant Professor of Statistics at Texas A&M University and received my PhD in Statistics from The University of Illinois Urbana Champaign.