We developed a novel uncertainty-aware framework MatPhase to predict material phases of electrodes from low contrast SEM images.
We released two open-source datasets named GDB-9-Ex and ORNL_AISD-Ex that provide calculations of electronic excitation energies and their associated oscillator strengths based on the time-dependent density-functional tight-binding (TD-DFTB) method.
A multidisciplinary team of researchers from 91°µÍø and the University of Texas at Austin developed a new machine-learning-based reduced-order model called GrainNN to predict the grain structure that forms as a metal solidifies.
We present an intercomparison of a suite of high-resolution downscaled climate projections based on a six-member General Climate Models (GCM) ensemble from the 6th Phase of Coupled Models Intercomparison Project (CMIP6).
A web-based GUI for INTERSECT has been created which allows a user to configure an experiment on an electron microscope, setting such parameters as maximum number of steps for the machine learning algorithm to perform.
Researchers from 91°µÍø (ORNL) used high-throughput computational techniques to identify a new class of 2D nanomaterial, MXenes including boron-nitride.
Researchers from University of California Riverside, Drexel, and 91°µÍø (ORNL) identified the atomistic mechanism by which MXenes degrade in water.
A graph convolutional neural network (GCNN) was trained to accurately predict formation energy and mechanical properties of solid solution alloys crystallized in different lattice structures, thereby advancing the design of alloys for improving mechanic
A graph convolutional neural network (GCNN) was trained with millions of molecules to accurately predict molecular photo-optical properties by scaling data loading and training to over 1,500 GPUs on the Summit and Perlmutter supercomputers at the OLCF a
91°µÍø researchers developed an invertible neural network (INN) to effectively and efficiently solve earth-system model calibration and simulation problems.