Robust Assessment of Surrogate Model Credibility
Abstract not provided.
Publications
Search results
Jump to search filtersUnraveling Sensitivity and Ensuring Reliability in Reynolds Stress Predictions for Data-Driven RANS
Abstract not provided.
CREDIBILITY ASSESSMENT OF MACHINE LEARNING-BASED SURROGATE MODEL PREDICTIONS ON NACA 0012 AIRFOIL FLOW
Proceedings of 2024 Verification, Validation, and Uncertainty Quantification Symposium, VVUQ 2024
The use of surrogate models in computational mechanics is an area of high interest due to the potential for significant savings in computational cost. However, assessment and presentation of evidence for surrogate model credibility has yet to reach a standard form. The present study utilizes a deep neural network as a surrogate for a computational fluid dynamics simulation in order to predict the coefficients of lift and drag on a NACA 0012 airfoil for various Reynolds numbers and angles of attack. Using best practices, the credibility of the underlying simulation predictions and of the surrogate model predictions are analyzed. Conclusions are drawn which should better inform future uses of surrogate models in the context of their credibility.
erification and Validation Using the Null Hypothesis as a Philosophical Grounding
Abstract not provided.
Data Credibility in Scientific Machine Learning
Abstract not provided.
Quantifying Uncertainty in Machine Learning Models for Time Series Classification
Abstract not provided.
Finding Confidence and Meaning in Verification
Abstract not provided.
Exploring the consequences and lessons from Underspecification in ML models
Abstract not provided.
Exploring the consequences and lessons from Underspecification in ML models (MLDL submission)
Abstract not provided.
V&V Adjacent Activities
Abstract not provided.
Sandia ATDM: Performance and Portability Accomplishments
Abstract not provided.
Sandia ATDM Program: Application Productivity and Performance Portability
Abstract not provided.
The Power and Secrets of the Sigmoid Function: A Numerical Swiss Army Knife
Abstract not provided.
ECP External Review - Sandia National Laboratories ATDM (WBS X.X.X.X)(KPP-2)
Accomplishments to date are provided for several programs: SPARC for Virtual Flight Testing; EMPIRE for Electromagnetic Plasma Physics; and the Science Challenge Problem Description.
FY19Q2 report for ATDM AD projects to ECP
Performance to Plan highlights are provided.
Scientific Method and Computational Science: Friends or Enemies?
Abstract not provided.
Validation Assessment of Hypersonic Double-Cone Flow Simulations using UQ Sensitivity Analysis and Validation Metrics
Abstract not provided.
Validation Assessment of Hypersonic Double-Cone Flow Simulations using Uncertainty Quantification Sensitivity Analysis and Validation Metrics
Abstract not provided.
Expansion Capturing Methods
Abstract not provided.
Verification for Hypersonic Reacting Turbulent Flow
Abstract not provided.
The Default Uncertainty is ZERO
Abstract not provided.
A Validation Study for a Hypersonic Flow Model
Abstract not provided.
ASC/ATDM FY18 L2 Milestone Midyear Review
Abstract not provided.
FY18 L2 Milestone: SPARC Initial Review
Abstract not provided.
Implica(ons of the Non-Zero Limit for Entropy Genera(on in the Inviscid Limit Solu(ons of Nonlinear Conserva(on Laws
Abstract not provided.
Results 1–25 of 108