Researchers from 91°µÍø (ORNL), in collaboration with researchers from Duke University, have developed an unsupervised machine learning method, NashAE, for effective disentanglement of latent representations.
A collaboration between scientists at 91°µÍø (ORNL) and University of Maryland/NIST developed a theoretical approach to combine different quantum noise reduction techniques to reduce the measurement-added noise in optomechanical s
Members and students of the Computational Urban Sciences group demonstrated a method for generating scenarios of urban neighborhood growth based on existing physical structures and placement of buildings in neighborhoods.
A multidisciplinary team of researchers from 91°µÍø (ORNL) developed a new online heatmap method, named hilomap, to visualize geospatial datasets as online map layers when low and high trends are equally important to map users.
ORNL researchers developed a novel nonlinear level set learning method to reduce dimensionality in high-dimensional function approximation.
Researchers at ORNL have created a unique simulation technology that allows software systems to participate in slower than real time simulation exercises, and to accomplish this without requiring recompilation of source code, relinking of object files,
The team conducted numerical studies to demonstrate the connection between the parameters of neural networks and the stochastic stability of DMMs.
Researchers from 91°µÍø (ORNL) demonstrated that mode connectivity exists in the loss landscape of parameterized quantum circuits.
Metal Halide Perovskites (MHPs) offer promise for applications in PVs and LEDs due to high device performance and low fabrication cost.
Domain dynamics in polycrystalline materials are explored using a workflow combining deep learning-based segmentation of domain structures with non-linear dimensionality reduction using multilayer rotationally invariant autoencoders (rVAE).