ORNL researchers developed a novel nonlinear level set learning method to reduce dimensionality in high-dimensional function approximation.
Filter Research Highlights
Area of Research
The team conducted numerical studies to demonstrate the connection between the parameters of neural networks and the stochastic stability of DMMs.
Quantum Monte Carlo simulations reveal that Cooper pairs in the cuprate high-Tc superconductors are composed of electron holes on the Cu-d orbital and on the bonding molecular orbital constructed from the four surrounding O-p orbitals.
Generative machine learning models, including GANs (Generative Adversarial Networks), are a powerful tool toward searching chemical space for desired functionalities.
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.
Researchers built a deep neural network to estimate the compressibility of scientific data.