Opening Workshop: August 20 – 24, 2018

Location

This workshop will be held at Gross Hall on the campus of Duke University.

Description

The SAMSI program on Model Uncertainty: Mathematical and Statistical brings statisticians and applied mathematicians together with disciplinary scientists from a variety of fields, to better understand the effects of modeling and uncertainty on predictions. This workshop provides the foundation of the MUMS year, examining the theoretical basis for statistical uncertainty, the strengths and weaknesses of models of real world processes and the uncertainty in those processes, computational methods to solve model equations, and the degree of confidence in predictions and inferences resulting from the analysis.

Much of scientific activity of this program will arise from Working Groups – groups of scientists interested in a common theme, who will meet regularly during the year. On Thursday afternoon, a series of brief presentations will be scheduled, to seed the formation of Working Groups. Organizational meetings of these proposed groups will follow, and continue into Friday morning. Among the potential topics of Working Groups are: Foundations of Statistical Model Uncertainty; Modeling Across Scales; Materials Informatics and Mechanics; Reduced Order Models; Uncertainty in Extrapolative Settings; Biomedical Data and Precision Medicine (joint with the PMED program); Stochastic Discretization; Uncertainty in Geoscience; the Small Data problem; Uncertainty and Machine Learning. Multiple Working Groups on similar topics may be formed.

** Planning for this workshop is ongoing. As more information becomes available, it will be updated here **


Tentative Schedule and Supporting Media

Printable Schedule
Speaker Abstracts
Poster Session Titles

Confirmed Speakers currently include:

Monday, August 20, 2018
Gross Hall, Duke University, Durham, NC
** DAY 1 Coverage STREAMING LIVE at: https://youtu.be/wKRamMOpYIk **

Time Description Speaker Slides Videos
8:30am Registration
8:50-9:00am Welcome and Introductory Information
Overview Lectures:
9:00-10:00am Model Uncertainty and Uncertainty Quantification Merlise Clyde, Duke University
10:00-10:30am BREAK
10:30-11:30am Principles of Predictive Computational Science: Predictive Models of Random Heterogeneous Materials and Tumor Growth Tinsley Oden, University of Texas
11:30am-12:30pm An Overview of Reduced-Order Models and Emulators Elaine Spiller, Marquette University
12:30-1:30pm LUNCH on own
Theoretical Foundations of Model Uncertainty:
1:30-2:30pm Hierarchical Bayesian Models for Inverse Problems and Uncertainty Quantification Bani Mallick, Texas A&M University
2:30-3:30pm On the Impact(s) of Structural Model Error on Simulation Modelling Leonard Smith, London School of Economics, Pembroke College, Oxford
3:30-4:00pm BREAK
4:00-5:00pm Quantifying Nonparametric Modeling Uncertainty with BART Edward George, Wharton, University of Pennsylvania
5:00-7:00pm Poster Session and Reception

Tuesday, August 21, 2018
Gross Hall, Duke University, Durham, NC

Time Description Speaker Slides Videos
9:00-10:00am The Isaac Newton Institute Uncertainty Quantification Programme: A Personal Perspective Peter Challenor, University of Exetor
10:00-10:30am BREAK
10:30am-12:00pm Panel on Calibration in the Face of Model Discrepancy Matthew Plumlee, Northwestern University
Mengyang Gu, Johns Hopkins
Georgios Karagiannis, University of Durham
12:00-1:00pm LUNCH
Model Reduction:
1:00-2:30pm Machine-Learning Error Models for Quantifying the Epistemic Uncertainty in Low-Fidelity Models Kevin Carlberg, Sandia National Laboratories
2:30-3:30pm Emulators for models and Complexity Reduction Akil Narayan, University of Utah
3:30-4:00pm BREAK
4:00-5:00pm Data-Driven Discovery of Governing Physical Laws and their Parametric Dependencies in Engineering, Physics and Biology Nathan Kutz, University of Washington

Wednesday, August 22, 2018
Gross Hall, Duke University, Durham, NC

Time Description Speaker Slides Videos
Extrapolation:
9:00-10:00am Extrapolation: The Art of Connecting Model-Based Predictions to Reality David Higdon, Virginia Tech
10:00-10:30am BREAK
10:30am-11:30am Bound-to-Bound-Data-Collaboration: Prediction on the Feasible Set Michael Frenklach, University of California, Berkeley
11:30am-12:30pm Model Discrepancy and Physical Parameters in Calibration and Prediction of Computer Models Jenny Brynjarsdóttir, Case Western Reserve University
12:30-1:30pm LUNCH
Materials:
1:30-2:30pm Modeling and Algorithmic Aspects of UQ for Material with Multiscale Behavior Roger Ghanem, University of Southern California
2:30-3:30pm Materials Innovation Driven by Data and Knowledge Systems Surya Kalidindi, Georgia Institute of Technology
3:30-4:00pm BREAK
4:00-5:00pm Panel on Materials Laura Swiler, Sandia National Laboratories
Michael Demkowicz, Texas A&M University
Ralph Smith, NC State University
5:00pm MUMS Workshop Social Event Event sponsored by NC Chapter of ASA

Thursday, August 23, 2018
Gross Hall, Duke University, Durham, NC

Time Description Speaker Slides Videos
Model and Data Fusion:
9:00-10:00am UQ Data Fusion: An Introduction and Case Study Robert Wolpert, Duke University
10:00-10:30am BREAK
10:30-11:30am Amy Braverman, JPL/Caltech
11:30am-12:30pm Inferring Release Characteristics from an Atmospheric Dispersion Model using Bayesian Adaptive Splines Bruno Sanso, University of California, Santa Cruz
12:30-1:30pm LUNCH on own
1:30-2:30pm Working Groups Overview/Proposals
2:30-3:00pm BREAK
3:00-5:00pm Working Groups Activity Rooms:  304B, 318, 324, 359, 107

Friday, August 24, 2018
Gross Hall, Duke University, Durham, NC

Time Description Speaker Slides Videos
9:00-10:00am Working Group Activity Rooms:  304B, 318, 324, 359, 107
10:00-10:30am BREAK
10:30am-12:00pm Working Groups Finalized
12:00-1:00pm LUNCH on own
1:00pm Shuttle to RDU Airport

Questions: email mums@samsi.info