Invited Paper Session

Statistical and Machine-Learning Innovations for Understanding Climate Extremes and Uncertainty

Thu, Aug 6, 8:30 AM - 10:20 AM Room CC-253B Thomas M. Menino Convention & Exhibition Center

Matthias KatzfussOrganizerMatthew HeatonChair

Section on Statistics and the Environment co: Uncertainty Quantification in Complex Systems Interest Groupco: Section on Physical and Engineering Sciences Applied

3 Presentations

Distributional climate-model calibration using variational autoregressive Gaussian processes

9:25 AM - 9:50 AM

Matthias Katzfuss (University of Wisconsin–Madison)

Bayesian transport map Bayesian optimization Spatial statistics Surrogate modeling Simulation-based inference Likelihood-free inference

Flow-Based Conformal Predictive Distributions

9:00 AM - 9:25 AM

Trevor Harris (University of Connecticut)

Conformal Prediction Conformal Inference Uncertainty Quantification Generative Modeling High Dimensional Climate

Studying Teleconnection in Climate Extremes using Statistical and Deep Learning Approaches

8:35 AM - 9:00 AM

Bo Li (Washington University in St. Louis)

Co-authors: shen yan (University of Illinois, Urbana-Champaign), Trevor Harris (University of Connecticut), Upmanu Lall (Columbia Univ)

Discussant

Christopher Wikle (University of Missouri)