SEARCH · Search NASA

Results for “Learning Algorithm”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 937 records · Page 52

All sources NASA NTRS DOE OSTI

DOE OSTI2018

Non-Traditional Supercomputing

The High-Performance Computing of today is much different than the supercomputing resources a few decades ago. The supercomputers of the past were huge complexes of hundreds of interconnected computers that had large teams of specialists to keep them running. Today, while supercomputers like Summit (the world's most powerful supercomputer as of June 2018, installed at Oak Ridge National Laboratory) are still in high demand for extremely fast computational power (over 200,000 trillion calculations per second, or 200 petaflops), researchers at Sandia now have access to non-traditional computing resources in the way of very powerful graphics processing units (GPUs). These GPU systems have enabled an entirely new avenue of exploration for Sandia, allowing researchers to further tackle problems in energy, advanced materials, artificial intelligence, and nuclear weapons. Many teams at Sandia utilize these GPU clusters to build models through the use of machine learning. These models are faster representations of their code-driven counterparts and can often be leveraged from the exascale computing resources of traditionally larger systems with huge boosts in performance. We highlight two resources for these smaller systems: the team that builds the hardware, and the team that researches and builds the machine learning algorithms.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

Safeguards Technology Development Program Annual Report of FY 2018

In this project, Brookhaven National Laboratory (BNL) and Sandia National Laboratories (SNL) are jointly developing deep machine learning algorithms to improve the review process of surveillance data by identifying objects of interest in image-review software. The technique being developed in this project can be used to reduce the burden on International Atomic energy Agency (IAEA) safeguards inspectors during surveillance review.

36 MATERIALS SCIENCE↗

DOE OSTI2017

The Bird project: Using Big Data tools to support Search Analytics

The Bird project explored the use of big data analytics tool to improve the findability of information within the Sandia internal network. We were able to perform query classification utilizing the supervised learning algorithms in the Apache Spark library. By relying on the distributed processing capabilities provided by the Apache Hadoop framework, we successfully processed the large query log files needed to train the models in this effort. The capabilities developed in this project are being used to enhance the effectiveness of the enterprise search engine.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2019

Final Technical Report

This report highlights of the last 3 years research. Discovery that one-layer (1L) epitaxial graphene plus one carbon interfacial buffer layer (BfL) , both grown by sublimation on SiC(0001), present a stiffness upon indentation similar/larger than bulk diamond and resist to perforation by a diamond indenter. Discovery of an sp 2 - sp 3 phase transformation upon indentation at room temperature of 1L+BfL epitaxial graphene on SiC, the new ultra-hard, ultra-stiff and insulating sp 3 phase was named Diamene . Demonstration of Å-Indentation (ÅI) measurements to measure the stiffness (up to 1 TPa) of ultra-thin (down to 1 atomic layer) films on substrates. First measurement by ÅI of the stiffness of Q-Carbon nano-filaments. Discovery of an oscillatory behavior in friction forces and work function values for even and odd number of layers in polycrystalline MoS 2 films, associated to the oscillatory piezoelectricity of the grains . Demonstration that monolayer MoS 2 can efficiently screen the vdW interaction of the underlying substrate with external systems. Identification of SiC buried terraces and graphene number of layers in a continuous film of epitaxial graphene on SiC by machine learning algorithms. New understanding of defects formation in monolayer MoS 2 . Demonstration of a new method based on thermal scanning probe lithography (t-SPL) for patterning electrodes on 2D materials overcoming EBL limitations . Demonstration of t-SPL for patterning nanoscale magnetic quantum waveguides . New understanding of how charged nanoparticles electrostatically assemble in 1D .

77 NANOSCIENCE AND NANOTECHNOLOGY↗

DOE OSTI2019

Machine Learning for Solar Technology Portfolio Management

Technology readiness level (TRL) labeling and forecasting today is done entirely manually. A computationally-assisted model of a TRL-like classification for estimation and tracking could enable a research and development (R&D) portfolio manager to more accurately estimate the relative level of development for technologies in that portfolio. This report describes the work to understand readiness level systems, train a machine learning algorithm to recognize textual indicators of particular readiness levels within written material from carefully chosen case studies in renewable technology development, and apply that algorithm to predict the readiness levels of other technology development cases. We account for the results of each step of the work conducted, including the predictive accuracy of the trained model: 54% correct, within a +/-1 readiness level error threshold and a root mean squared error of 1.88. Great potential remains to improve and apply this model for management of individual projects and across portfolios.

14 SOLAR ENERGY↗

DOE OSTI2019

Machine Learning for Solar Technology Portfolio Management

14 SOLAR ENERGY↗

DOE OSTI2019

Understanding the Effects of Tapering on Gaussian Process Regression

Gaussian process regression is a computationally expensive machine learning algorithm that requires the solution of a system of linear equations with a dense covariance matrix. We propose a modification that uses tapering to introduce sparsity into the covariance matrix followed by matrix reordering schemes to reduce the bandwidth of the sparse matrix. We also apply iterative refinement to our approximate solution in order to recover some of the accuracy lost to tapering. Our numerical tests showed that this modification works best with datasets that have many small values in the covariance matrix. This restricts the usefulness of the modification, so other techniques to reduce computational time may be worth considering in the future.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

An Innovative HIV “Mosaic” Vaccine Design Based on Machine Learning

Approximately 37 million people worldwide have the human immunodeficiency virus (HIV). Because of HIV’s incredible diversity — each infected person carries a unique set of viruses — an effective vaccine against HIV needs to stimulate an immune response that can recognize its many variants. Using a machine-learning algorithm and the NIH funded Los Alamos National Laboratory database of hundreds of thousands of HIV sequences gathered from around the globe, Los Alamos scientists were able to design small sets of sequences that in combination provide optimal coverage of HIV’s diversity. The mosaic vaccine is the world’s first computationally-designed vaccine. Through a decade of testing it has continued to show promise, and so currently is being tested in two human efficacy trials to evaluate whether the investigational vaccine regimen can reduce the incidence of HIV infection among 2,600 women in sub-Saharan Africa, and 3,800 men who have sex with men and transgender people in the Americas.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2020

YOLO Test Software v1.4

This is a user manual of the software being developed at Brookhaven National Laboratory to test deep learnings algorithms. (YOLO) You Only Look Once. V1.4 added new features, including Display metric average recall, early termination option and display filename/image.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

Group Structure Machine Learning Proposal

Nuclear data is the linchpin underwriting several fundamental capabilities and mission needs at LANL. New techniques such as machine learning can be brought to bear to solve old problems such as multigroup cross-section accuracy. In neutron transport, generating multigroup cross sections is a complex and arcane task, but a crucial one, as accurate solutions require appropriate cross sections. There are two key challenges when generating multigroup cross sections: (1) choosing an accurate weight function, and (2) choosing appropriate energy boundaries. Often, energy boundaries are chosen using “expert judgment” that is not documented and is difficult to replicate. The long-standing Los Alamos 30-group structure has been in use since at least 1969 and is still in use today. Simplistic attempts over the years since to improve on the 30-group structure have been met with limited success. Machine learning algorithms would enable the selection of appropriate, problem-dependent group boundaries without an inordinate investment of scientist time. We will develop workflows and tools to enable these improved group boundary choices, which will reduce uncertainty and increase predictive capability of neutron-transport applications at LANL.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2020

Cognition at the Point of Sensing

Over the last 15 years, compressive sensing techniques have been developed which have the potential to greatly reduce the amount of data collected by systems while preserving the amount of information obtained. A cost of this efficiency is that a computationally-intensive optimization routine must be used to put the sensed data into a form that a person can interpret. At the same time, machine learning techniques have experienced tremendous growth as well. Machines have demonstrated the ability learn how to effectively perform tasks such as detection and classification at speeds much faster than humanly possible. Our goal in this project was to study the feasibility of using compressive sensing systems "at the edge." That is, how can compressive sensing sensors be deployed such that information is created at the remote sensor rather than sending raw data to a central processing location? Studies were performed to analyze whether machine learning could be done on the compressively sensed data in its raw form. If a machine is performing the task, is it possible to do so without putting the data into a human interpretable form? We show that this is possible for some systems, in particular a compressive sensing snapshot imaging spectrometer. Machine learning tasks were demonstrated to be more effective and more robust to noise when the machine learning algorithm worked on data in its raw form. This system is shown to outperform a traditional spectrometer. Techniques for reducing the complexity of the reconstruction routine were also analyzed. Techniques for such as data regularization, deep neural networks, and matrix completion were studied and shown to have benefits over traditional reconstruction techniques. In this project we showed that compressive sensing sensors are indeed feasible at the edge. As always, sensors and algorithms must be carefully tuned to work in the constrained environment. In this project we developed tools and techniques to enable those analyses.

47 OTHER INSTRUMENTATION↗

DOE OSTI2020

Applying Machine Learning to the Classification of DC-DC Converters. NA-22 Final Report

Tools are now available that enable measurement electromagnetic radiation (EMR) from active electronics in an item. This radiation may be intended WIFI or cellular network links, for example or unintended such as the switching noise generated by DC-to-DC converters. It would be extremely valuable to have the capability to discriminate between the low-voltage DC-to-DC converters or other digital noise prevalent in most modern electronics, versus the high-voltage DC-to-DC converters used in utility firesets. Previous work performed under a Sandia Laboratory Directed Research and Development (LDRD) project on Charge State Detection using a deep neural network has been continued in this effort. A state-of-the-art supervised machine learning algorithm has not only been extended to discriminate between low and high voltage converters but has been validated in determining a converters make and model.

42 ENGINEERING↗

DOE OSTI2021

Feature Detection

Focal Area(s): This proposal aims to develop and evaluate statistical models and machine learning algorithms for detecting and tracking features in spatiotemporal remotely sensed data with uncertainty quantification. We focus a particular application on the detection of sea ice leads and ridges in the Arctic and use these key sea ice features for model calibration and to gain insight into the physics of sea ice thermodynamics and deformation.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Constraints on Neutrino Oscillation Parameters from Neutrinos and Antineutrinos with Machine Learning

NOvA is a two detector, long baseline neutrino oscillation experiment measuring the oscillations of muon neutrinos from the \numi neutrino beam over a baseline of \SI{810}{km}. The experiment uses four oscillation channels, $\numu \rightarrow \numu$, $\numubar \rightarrow \numubar$, $\numu \rightarrow \nue$, and $\numubar \rightarrow \nuebar$, with a peak neutrino energy of \SI{1.8}{GeV}. This dissertation describes the analysis of these channels using a dataset of $13.6\times10^{20}$ protons on target neutrino beam mode and $12.5\times10^{20}$ protons on target antineutrino beam mode. The analysis makes use of improvements in the treatment of systematic uncertainties and machine learning techniques to reconstruct neutrino interactions. A technique for decorrelating systematic errors using principle component analysis was utilized to reduce and optimize neutrino cross section and beam related uncertainties. The improved machine learning algorithms make use of convolutional ne ural net works for neutrino event classification, particle classification, and instance segmentation. The selection of neutrino signal events utilizing the neutrino event classifier shows an efficiency of 63\% for the selection of electron neutrinos in neutrino beam mode and 75\% for electron antineutrinos in antineutrino beam mode. Using this algorithm, 82 appearing electron neutrino candidates and 33 appearing electron antineutrino candidates were observed with expected backgrounds of 26.8 and 14.0 respectively. In addition, 211 surviving muon neutrino candidates and 105 muon antineutrino candidates were identified with a purity of more than 96\% using the same neutrino event classifier. Fitting these data to the three flavor neutrino oscillation model, using constraints on \thetaonetwo, \thetaonethree, and \dmsqonetwo from solar and reactor neutrino experiments, the oscillation parameters are measured to be $\sintwothree = 0.57^{+0.04}_{-0.03}$, $\dmsqthreetwo = \SI[parse-numbers= false]{+ 2.41\pm0.07 \times 10^{-3}}{eV^2}$, and $\dcp=0.82^{+0.27}_{-0.87}\pi$ with a preference for the normal neutrino mass hierarchy. Leading systematic uncertainties for these measurements come from detector calibration and neutrino interaction models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2021

Boundary Layer Climatology at ARM Southern Great Plains

Operational since 1992, the Atmospheric Radiation Measurement (ARM) southern great plains (SGP) site at Oklahoma, USA has become a reference research site for meteorological studies. Due to an open data policy the ARM data are used by researchers all over the world. In this report, we review the long-term climatology of the atmospheric boundary layer, SGP instrumentations, the site and some site-specific atmospheric conditions which potentially effect wind turbines within the region. As the atmospheric boundary layer is bounded and influenced by the surface, observations of surface radiation components and heat fluxes are crucial in understanding land-atmosphere interactions. The entrainment of air, updrafts, downdrafts and boundary layer height characteristics is needed for understanding the structure and growth of the atmospheric boundary layer. Therefore, measurements from both surface in-situ and remote sensing observations at SGP provide an overall climatology and their interactions from surface up to the boundary layer. Measurements from a 60 m meteorological tower, surface flux stations, disdrometers, soil temperature and moisture flux plates, coherent Doppler lidar, Raman lidar, radiosondes, and satellite data at SGP central facility were analyzed. All the measurements were generally within a few square kilometers of each other at the central facility. This report focuses on data from January 2010 to June 2020 at SGP central facility. The various sections describe the ARM SGP site and surrounding wind turbines; in-situ and remote sensing instrumentation used in the report; provides mathematical equations to analyze fluxes, turbulence and other boundary layer parameters; a climatological analysis of surface winds, fluxes and thermodynamic parameters for several years; an analysis of observed winds in the framework of Monin-Obukhov Similarity Theory; an analysis of the boundary layer winds and direction from a Doppler lidar; multi-year turbulence estimates through the boundary layer from a Doppler lidar; atmospheric boundary layer water vapor and relative humidity profiles from Raman lidar; cloud base height and boundary layer height from multiple sensors and satellite data; and finally site specific atmospheric conditions, such as nocturnal low-level jets. Diurnal, seasonal and yearly variations of surface, sub-surface and boundary layer quantities, such as wind speed, direction, temperature, atmospheric stability, soil temperature, and various atmospheric fluxes at SGP showed distinct trends useful for focused modeling studies. The applicability of surface similarity theory on ARM SGP data is also evaluated, which showed northerly flows are aligned with MO theory estimates compared to southerly flows. Boundary layer winds and direction profiles for several years from a Doppler lidar shows a consistent presence of a nocturnal low-level jet and predominant southerly wind directions through the boundary layer at SGP. The inter-annual variability at SGP is low (<3.5%), with a mean annual wind speed of approximately 7 m s -1 at 100 m above ground level. Boundary layer turbulence and moisture transport from Doppler and Raman lidars are evaluated, which provides evidence of increased water vapor mass flux into the great plains during nocturnal low-level jets. The moisture flux from nocturnal low-level jets is observed to be maximum during summer periods. A novel machine learning algorithm is implemented to accurately estimate the planetary boundary layer height, providing further insights into growth and destruction of the convective boundary layer height during various seasons and land-atmosphere conditions. The high frequency of low-level clouds during winter, spring and fall seasons is validated using the multi-sensor array and satellite estimates of cloud top height. Satellite vegetative fraction data provides insight into seasonal surface roughness and vegetation variability around SGP site.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

CIS Project 22359, Final Technical Report. Discretized Posterior Approximation in High Dimensions

Our primary aim in this work is to understand how to efficiently obtain reliable uncertainty quantification in automatic learning algorithms with limited training datasets. Standard approaches rely on cross-validation to tune hyper parameters. Unfortunately, when our datasets are too small, holdout datasets become unreliable—albeit unbiased—measures of prediction quality due to the lack of adequate sample size. We should not place confidence in holdout estimators under conditions wherein the sample variance is both large and unknown. More poigniantly, our training experiments on limited data (Duersch and Catanach, 2021) show that even if we could improve estimator quality under these conditions, the typical training trajectory may never even encounter generalizable models.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

Deep Learning Based Approach to Integrate MyShake's Trigger Data with ShakeAlert for Faster and Robust EEW Alerts

Earthquake Early Warning (EEW) systems depend on the dense seismic network to make fast and accurate earthquake detections to issue useful early alerts. The recent development and implementation of the ShakeAlert system is relying on the west coast seismic network to detect and estimate the earthquake parameters in real-time. While working on making improvements on the current system, one potential approach is to include more seismic recordings from various sources, such as the low-cost smartphone seismic network, such as the MyShake network, or the Google Android smartphones, since these smartphone seismic networks have much more portable stations (phones) to potentially provide useful data to the system. This report summarizes the initial exploration of using a deep learning approach to combine the data from both traditional seismic stations and the smartphone data in California. Due to the heterogeneity nature of the data, we aggregate data from the phones (using simulation data), as well as that from traditional seismic stations to grid cells. By generating real-time triggering ratio grid cell maps, the designed deep learning algorithm can process the data from multiple sources and detect the earthquake faster than only using that from a traditional seismic network.

58 GEOSCIENCES↗

DOE OSTI2022

Power System Waveform Classification Using Time-Frequency and CNN

Many modern reclosers and circuit breakers have microprocessor relays that record waveforms of system events. In some cases, utilities may record a half-a-dozen event captures for every event. This is thousands of events per year. The industry needs faster, more automated, more conclusive, and easy-to-use systems that can process massive amounts of event recordings without extensive input/support from power system engineers. To address the need for a commercially viable solution that can classify waveform data, energies were directed to develop a universal neural network (NN) structure (deep learning algorithm) that works for a wide variety of system event types. The structure that showed the most promise was one that included the use of spectrograms. The technique has shown positive results in audio engineering, particularly with respect to speech recognition. A waveform signature could be treated as a spoken word like audio waveforms for specific things such as “YES” or “UP”. No two people produce the exact same waveform when speaking each of these words, but audio processing algorithms based on spectrograms and convolutional neural networks (CNN) can still distinguish the word regardless of the speaker. No two circuits produce the exact same waveform for a given event, but the NN can be trained to classify the event type regardless of the circuit or location on the circuit. A Power System Neural Network (PSNN) has been developed to use a CNN to classify events within waveform data for power systems. The waveform is converted to an array of values by way of spectrograms and interpreted as an image. This image is passed into the CNN. The test results on independent simulated test and validation datasets show greater than 99% accuracy. While the results thus far are based on simulated data, the performance of the PSNN is very promising and should work for a wide variety of power system conditions of interest. Ultimately, much of the custom code and tools used today and much of the manual effort expended today may be automated using this PSNN.

24 POWER TRANSMISSION AND DISTRIBUTION↗