Conference Program Home
  My Program

All Times EDT

Legend:
* = applied session       ! = JSM meeting theme

Activity Details

403 * ! Wed, 8/10/2022, 10:30 AM - 12:20 PM CC-151A
Research Advances at the Interface of Uncertainty Quantification and Machine Learning for High-Consequence Problems — Invited Papers
Section on Statistics in Defense and National Security, Section on Statistical Learning and Data Science, IEEE Computer Society
Organizer(s): Ahmad Rushdi, Stanford University
Chair(s): Erin Acquesta, Sandia National Laboratories
10:35 AM Variational Inference with NoFAS: Normalizing Flow with Adaptive Surrogate for Computationally Expensive Models
Yu Wang, Notre Dame University; Daniele Schiavazzi, University of Notre Dame; Fang Liu, Univerisity of Notre Dame
10:55 AM Assessing the Quality of Uncertainty Estimates in Deep Learning
Jason Adams, Sandia National Laboratories
11:15 AM Extreme Learning Machines for Variance-Based Global Sensitivity Analysis
John Darges, North Carolina State University; Alen Alexanderian, North Carolina State University; Pierre Gremaud, North Carolina State University
11:35 AM Efficient Variational Approach to Sparse BNN for Model Compression
Diptarka Saha, University of Illinois, Urbana-Champaign; Feng Liang, University of Illinois, Urbana-Champaign ; Zihe Liu, University of Illinois, Urbana-Champaign
11:55 AM Discussant: Daniel Ries, Sandia National Labs
12:15 AM Floor Discussion