We successfully utilized OCLF ORNL GPU computing resources for efficient uncertainty analysis, which addressed the computational overhead caused by our proposed probabilistic models.
In this work we focus on dynamics problems described by waves, i.e. by hyperbolic partial differential equations.
A research team from ORNL, Pacific Northwest National Laboratory, and Arizona State University has developed a novel method to detect out-of-distribution (OOD) samples in continual learning without forgetting the learned knowledge of preceding tasks.
ORNL researchers developed a novel nonlinear level set learning method to reduce dimensionality in high-dimensional function approximation.
The team conducted numerical studies to demonstrate the connection between the parameters of neural networks and the stochastic stability of DMMs.
Estimating complex, non-linear model states and parameters from uncertain systems of equations and noisy observation data with current filtering methods is a key challenge in mathematical modeling.
ORNL researchers developed a stochastic approximate gradient ascent method to reduce posterior uncertainty in Bayesian experimental design involving implicit models.
The researchers from ORNL have developed a new and faster algorithm for the graph all-pair shortest-path (APSP) problem.