We successfully utilized OCLF ORNL GPU computing resources for efficient uncertainty analysis, which addressed the computational overhead caused by our proposed probabilistic models.
We propose a novel deep learning method that achieves 170X average speed up compared to the original probabilistic marching cubes algorithm [1] implementation and performs predictions with an accuracy comparable to the original algorithm.
We propose the application of various visualization techniques, such as probability maps, confidence maps, level sets, and topology-based visualizations, to effectively communicate the uncertainty in source localization with clinicians.
A multidisciplinary team of researchers has developed an adaptive physics refinement (APR) technique to effectively model cancer cell transport.
A multi-university team first reported a unique lead-free ferroelectric compound - (Ca,Sr)3Mn2O7, which belongs to a class of materials described as hybrid improper ferroelectrics.
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.
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).
Researchers at Oak Ridge National Laboratory developed a new parallel performance portable algorithm for solving the Euclidean minimum spanning tree problem (EMST), capable of processing tens of millions of data points a second.
A team of Oak Ridge National Laboratory (ORNL) scientists involved in research topics of cybersecurity, statistical approaches, control systems, and dynamical models, reported a basic approach to security of physical systems that are interfaced with IT
A new file format, BP5, and accompanying serialization class has been developed in the ADaptable I/O System (ADIOS) framework.