Filter Research Highlights
Area of Research
![While simulating the ten-dimensional Lorentz 96 system, the Direct Filter method accurately and quickly captures the system’s parameters in comparison to the popular Ensemble Kalman filter (EnKF). Computer Science and Mathematics CSMD ORNL](/sites/default/files/styles/list_page_thumbnail/public/2021-05/direct_filter_method_for_parameter_estimation.png?h=1ee7460e&itok=ok8wz49e)
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.
![Radiofrequency pulse is an approach to control Rabi oscillation for quantum optics, MRI, etc. The ORNL method enables the ideal location of photon measurements in the pulse frequency and duration space to achieve a Rabi oscillation with desired Rabi and detuning frequencies. Left: the designed measurement points (from pink dots to red dots). Right: the evolution of the posterior distribution (the blue clouds) with the increase of measurement from 1 to 500. Computer Science and Mathematics CSMD ORNL](/sites/default/files/styles/list_page_thumbnail/public/2021-05/a_novel_method_for_bayesian_experimental_design_with_implicit_models.png?h=08c29538&itok=LmjpLXrG)
ORNL researchers developed a stochastic approximate gradient ascent method to reduce posterior uncertainty in Bayesian experimental design involving implicit models.
![The architecture of the Plexus resilient runtime system interfacing with programming model runtimes, libraries, system monitoring, and job and resource management. Computer Science and Mathematics Division CSMD ORNL](/sites/default/files/styles/list_page_thumbnail/public/2021-05/plexus-_a_pattern-oriented_runtime_system_architecture_for_resilient_extreme-scale_high-performance_computing_systems.png?h=8bd7d9f2&itok=AhEKp7Fp)
A team of researchers from Oak Ridge National Laboratory (ORNL) designed, implemented, and evaluated a high-performance computing (HPC) runtime system.
![(a) Overview of the new approach. (b) Comparison of EONS and proposed method on cart pole balancing task. Lines show average max fitness value at each epoch over ten evolutionary runs for each algorithm. Shaded areas show the one standard deviation range for the max fitness values across multiple evolutionary runs at each epoch. Computer Science and Mathematics Division CSMD ORNL](/sites/default/files/styles/list_page_thumbnail/public/2021-05/neuroevolution_of_spiking_neural_networks_using_compositional_pattern_producing_networks.png?h=23c51eb9&itok=lE83UYyn)
Researchers from Oak Ridge National Laboratory and the University of Central Florida have extended an evolutionary approach for training spiking neural networks.