Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed training”

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.

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

At least 487 records · Page 27

Machine learning to classify animal species in camera trap images: Applications in ecology

Motion-activated cameras (“camera traps”) are increasingly used in ecological and management studies for remotely observing wildlife and are amongst the most powerful tools for wildlife research. However, studies involving camera traps result in millions of images that need to be analysed, typically by visually observing each image, in order to extract data that can be used in ecological analyses. We trained machine learning models using convolutional neural networks with the ResNet-18 architecture and 3,367,383 images to automatically classify wildlife species from camera trap images obtained from five states across the United States. We tested our model on an independent subset of images not seen during training from the United States and on an out-of-sample (or “out-of-distribution” in the machine learning literature) dataset of ungulate images from Canada. We also tested the ability of our model to distinguish empty images from those with animals in another out-of-sample dataset from Tanzania, containing a faunal community that was novel to the model. The trained model classified approximately 2,000 images per minute on a laptop computer with 16 gigabytes of RAM. The trained model achieved 98% accuracy at identifying species in the United States, the highest accuracy of such a model to date. Out-of-sample validation from Canada achieved 82% accuracy and correctly identified 94% of images containing an animal in the dataset from Tanzania. We provide an r package (Machine Learning for Wildlife Image Classification) that allows the users to (a) use the trained model presented here and (b) train their own model using classified images of wildlife from their studies. The use of machine learning to rapidly and accurately classify wildlife in camera trap images can facilitate non-invasive sampling designs in ecological studies by reducing the burden of manually analysing images. Our r package makes these methods accessible to ecologists.

54 ENVIRONMENTAL SCIENCES↗

MO-G-201-04: Knowledge-Based Planning for Single-Isocenter Stereotactic Radiosurgery to Multiple Brain Metastases

Purpose: Single-isocenter, linac-based SRS for multiple brain metastases (multi-mets) can deliver highly conformal radiation doses and reduce overall patient treatment time compared to other therapy techniques. This study aims to quantify the dosimetric benefits of knowledge-based planning (KBP) for multi-met treatments. Methods: Using a previously-published KBP methodology (an artificial neural network (ANN) trained on single-target linac-based SRS plans), 3D dose distribution predictions for multi-met patients were obtained by treating each brain lesion as a solitary target and subsequently combining individual predictions into a single distribution using a dose-weighted geometric averaging to obtain the best results in the inter-target space. 17 previously-treated multi-met plans, with target numbers ranging from N=2–5, were used to validate the ANN predictions and subsequent KBP auto-planning routine. The fully-deliverable KBP plans were developed by converting dose distribution predictions into patient-specific optimization objectives while maintaining identical target normalizations (typically PTV V100%=D98%). Plan quality improvements were quantified by the difference between SRS quality metrics (QMs): δdQM=QM(clinical)-QM(KBP). QMs of interest were: gradient measure (GM), conformity index (CI), brain V10 and V5, brainstem D0.1cc and heterogeneity index (HI). Finally, overall plan quality was judged via blinded plan comparison by SRS-specializing physicians. Results: Two clinical plans were found to be significant outliers wherein plan quality was dramatically worse than KBP. Despite indicating KBP superiority, these were removed from the QM analysis to prevent skewing the results. In the remaining cases, clinical and KBP QMs were nearly identical with modest improvements in the KBP sample: δGM=0.12±0.56mm, δCI=−0.01±0.04, Brain δV10=0.8±2.6cc, brain δV5=6.3 ±10.7cc, brainstem δD0.1cc=0.06±1.19Gy and δHI= −0.04±0.05. Ultimately, 13/17 KBP plans were deemed superior to the manual plans in blinded physician review. Conclusion: The results demonstrate that KBP-driven automated planning in linac-based single-isocenter treatments for multiple brain metastases is indistinguishable from, or even better than, traditional manual planning. J. Hattangadi: Research Grant; Varian Medical Systems; K.L. Moore: Research Grant: Varian Medical Systems.

60 APPLIED LIFE SCIENCES↗

LTAU-FF: Loss Trajectory Analysis for Uncertainty in atomistic Force Fields

LTAU (Loss Trajectory Analysis for Uncertainty) is a technique for estimating the uncertainty in a model's predictions for any regression task by approximating the cumulative distribution function (CDF) of the model's errors over the course of training. The approximated CDF, combined with a similarity search algorithm in the model's descriptor space, can then be used to estimate the likelihood that the model's predictions on any given test point will fall below a chosen threshold. LTAU-FF (LTAU in atomistic Force Fields) is the application of LTAU specifically for use with atomistic force fields.

Vita, JoshuaA↗

Colorado Alternatively Powered Aircraft Airport Infrastructure Study

Advanced Air Mobility encompasses emerging technologies in propulsion of general aviation and commuter aircraft for the purpose of moving passengers and goods. The State of Colorado has a strong aerospace and aviation legacy, and this research explores the potential energy implications of a transition from traditional aircraft propulsion systems to more-electric, full electric and hydrogen powered aircraft in flight schools, and small community air service utilizing NASA's Regional Air Mobility Model. Effort includes identifying potential energy needs and cost-effective solutions to support those emerging needs analyzing potential benefits on on-site generation and storage to optimize energy systems.Advanced Air Mobility encompasses emerging technologies in propulsion of general aviation and commuter aircraft for the purpose of moving passengers and goods. The State of Colorado has a strong aerospace and aviation legacy, and this research explores the potential energy implications of a transition from traditional aircraft propulsion systems to more-electric, full electric and hydrogen powered aircraft in flight schools, and small community air service utilizing NASA's Regional Air Mobility Model. Effort includes identifying potential energy needs and cost-effective solutions to support those emerging needs analyzing potential benefits on on-site generation and storage to optimize energy systems.

33 ADVANCED PROPULSION SYSTEMS↗

Hofmann Stability Charts Revisited for PIP-II: From Classical Theory to Assumption-Free and ML-Driven Maps

The Hofmann stability chart remains a standard for visualizing parametric resonances in space-charge–dominated linacs, but its use typically relies on non-oscillatory Vlasov dispersion relations with simplifying assumptions (continuous focusing, KV phase space, linear optics, limited transverse–longitudinal coupling). We revisit the chart for the PIP-II linac along three tracks. (1) We reproduce the conventional maps in the (νz/νx, νx/ν0x) plane for relevant εz/εx, providing a validated reference. (2) We remove key assumptions by deriving stability surfaces directly from multi-particle tracking with realistic lattice discreteness, RF defocusing, solenoid/quad optics, and bunched-beam dynamics; local tunes and early-time growth rates are estimated from envelope oscillations and projected to the same coordinates. These assumption-reduced maps recover the canonical stopbands while revealing shifts and broadenings driven by tune modulation, non-KV distributions, and transverse–longitudinal coupling at PIP-II intensities. (3) We train a compact machine-learning surrogate that emulates the growth surface from zero-current optics, tune depression, emittance ratio, bunching factor, and selected lattice descriptors, enabling rapid scans and online working-point selection. We compare the three representations on representative PIP-II sections and discuss implications for commissioning guard bands, resonance avoidance, and routine operations.

Pathak, Abhishek [Fermilab] (ORCID:000000021704208↗

Sentinel protist taxa of the McMurdo Dry Valley lakes, Antarctica: a review

High-latitude meromictic lakes such as those in the Antarctic McMurdo Dry Valleys (MDV) harbor aquatic ecosystems dominated by the microbial loop. Within this habitat, which is limited year-round by light and nutrients, protists, or single celled eukaryotes, play outsized roles in the food web as the dominant primary producers and the apex predators. Thus, the MDV lake ecosystem represents an ideal system to study the role of sentinel protist taxa in carbon and nutrient cycling. The perennially ice-covered lakes are part of the McMurdo Long Term Ecological Research (McM LTER; mcmlter.org) established in 1993. In this review we will highlight the diversity and trophic roles of the MDV lake protist community and compare environmental factors driving spatiotemporal patterns in key protist taxa in two lakes within the McM LTER, Lakes Bonney and Fryxell. We will then discuss lessons learned from manipulated experiments on the impact of current and future climate-driven environmental change on sensitive protist taxa. Last, we will integrate knowledge gained from 25 years of lab-controlled experiments on key photosynthetic protists to extend our understanding of the function of these extremophiles within the MDV aquatic food webs. Our research group has studied the distribution and function of the MDV microbial community for nearly two decades, training the next generation of scientists to tackle future problems of these globally significant microbes. This review article will also highlight early career scientists who have contributed to this body of work and represent the future of scientific understanding in the Anthropocene.

54 ENVIRONMENTAL SCIENCES↗

Visualization for Classification in Deep Neural Networks

Recently, the techniques based on Deep Neural Networks (DNNs) have achieved a great performance in classification tasks in a wide range of applications, such as image recognition and natural language processing. However, DNN developers face a lot of trial and error during the development process and spend their efforts in debugging their network model through analyzing and understanding the classification results. As such, tools are needed that help the developers not only understand the results, but also suggest the ways to improve their model. In this paper, we propose a visual analytics tool for visualizing the classification results during the iterative development pipeline of a DNN model. Our tool enables exploring the classification results from any type of neural network models, identifying misclassified samples, examining the predicted score distributions of samples, and showing how the outcomes progressively change during the training process.

Chae, Junghoon↗

DIAGNOSING THE TIME DEPENDENCE OF ACTIVE REGION CORE HEATING FROM THE EMISSION MEASURE. II. NANOFLARE TRAINS

The time dependence of heating in solar active regions can be studied by analyzing the slope of the emission measure distribution coolward of the peak. In a previous study we showed that low-frequency heating can account for 0% to 77% of active region core emission measures. We now turn our attention to heating by a finite succession of impulsive events for which the timescale between events on a single magnetic strand is shorter than the cooling timescale. We refer to this scenario as a 'nanoflare train' and explore a parameter space of heating and coronal loop properties with a hydrodynamic model. Our conclusions are (1) nanoflare trains are consistent with 86% to 100% of observed active region cores when uncertainties in the atomic data are properly accounted for; (2) steeper slopes are found for larger values of the ratio of the train duration {Delta} {sub H} to the post-train cooling and draining timescale {Delta} {sub C}, where {Delta} {sub H} depends on the number of heating events, the event duration and the time interval between successive events ({tau} {sub C}); (3) {tau} {sub C} may be diagnosed from the width of the hot component of the emission measure provided that the temperature bins are much smaller than 0.1 dex; (4) the slope of the emission measure alone is not sufficient to provide information about any timescale associated with heating-the length and density of the heated structure must be measured for {Delta} {sub H} to be uniquely extracted from the ratio {Delta} {sub H}/{Delta} {sub C}.

79 ASTRONOMY AND ASTROPHYSICS↗

Need for research and training reactors for advanced reactor designs

Full text of publication follows. Research and training reactors have served a valuable role in helping train workforce for currently operating fleet of light water reactors. These research and training reactors have been used in reactor laboratory classes to familiarize the students with such vital concepts as approach to criticality, reactor period, neutron moderation, reactivity, flux distribution and leakage, etc. As the industry moves toward advanced non-light-water reactor designs, it is critical that research and training reactors be developed and deployed at university campuses to help train the new generation of nuclear and non-nuclear engineers who are likely to design, build, and operate these advanced reactors. Among the designs currently being pursued for nuclear power generation include molten salt, sodium cooled, and gas cooled designs, with options for various fuel forms. Thus, industry and DOE in collaboration with academic institutions should devise plans on how to familiarize the next generation of nuclear workforce with hands-on experience necessary for such designs. These research and training reactors will play a vital role in familiarizing the future workforce with hands-on experience with concepts associated with fast spectrum reactors, gas cooled reactors, and other features not associated with light water reactors. In addition to classical nuclear engineering concepts, these advanced research and training reactors can also be used for hands-on training as well as for research on features being considered in the design of GEN-IV reactors: cyber security for digital control room operations, hybrid energy system, hydrogen generation, district heating, autonomous control... (author)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

High–Resolution Maps of Near–Surface Permafrost for Three Watersheds on the Seward Peninsula, Alaska Derived From Machine Learning

Permafrost soils are a critical component of the global carbon cycle and are locally important because they regulate the hydrologic flux from uplands to rivers. Furthermore, degradation of permafrost soils causes land surface subsidence, damaging infrastructure that is crucial for local communities. Regional and hemispherical maps of permafrost are too coarse to resolve distributions at a scale relevant to assessments of infrastructure stability or to illuminate geomorphic impacts of permafrost thaw. Here we train machine learning models to generate meter–scale maps of near–surface permafrost for three watersheds in the discontinuous permafrost region. The models were trained using ground truth determinations of near–surface permafrost presence from measurements of soil temperature and electrical resistivity. We trained three classifiers: extremely randomized trees (ERTr), support vector machines (SVM), and an artificial neural network (ANN). Model uncertainty was determined using k–fold cross validation, and the modeled extents of near–surface permafrost were compared to the observed extents at each site. At–a–site near–surface permafrost distributions predicted by the ERTr produced the highest accuracy (70%–90%). However, the transferability of the ERTr to the sites outside of the training data set was poor, with accuracies ranging from 50% to 77%. The SVM and ANN models had lower accuracies for at–a–site prediction (70%–83%), yet they had greater accuracy when transferred to the non–training site (62%–78%). These models demonstrate the potential for integrating high–resolution spatial data and machine learning models to develop maps of near–surface permafrost extent at resolutions fine enough to assess infrastructure vulnerability and landscape morphology influenced by permafrost thaw.

54 ENVIRONMENTAL SCIENCES↗

Deep Spectroscopy with DESI for Photometric Redshift Training and Calibration

Deep spectroscopic samples can improve photometric redshift (photo-z) estimates and reduce uncertainties on redshift distributions. Such improvements can increase the cosmological constraining power of large imaging-based experiments such as the Vera C. Rubin Observatory’s Legacy Survey of Space and Time (LSST) and mitigate what may be a limiting systematic effect. We present results from the “DESI-Deep pilot” program, which was designed to assess the capability of the Dark Energy Spectroscopic Instrument (DESI) on the 4m Mayall telescope to measure redshifts of galaxies as faint as expected lensing samples for early LSST data (m i ≤ 24.5). We find that DESI is remarkably efficient at this task, with redshift success rates comparable to the results of observations from 10 m class telescopes with only ∼2 × longer integration time (rather than ∼8 × longer as would be expected from aperture-area scaling), while simultaneously achieving ∼30 times larger multiplexing. We also find that the signal-to-noise ratio of the spectra scales as expected for background-limited observations even for the longest exposure times (∼7 hr) and faintest targets in the program. These results demonstrate that DESI could provide the definitive redshift sample for the early years of LSST with a modest investment of observing time. Based upon the results of this program, we provide updated predictions for the time required to collect benchmark samples for photo-z training and calibration using a variety of spectroscopic facilities. Finally, we describe a potential “DESI-Deep” survey designed to train and calibrate photo-z’s for imaging experiments, and provide forecasts of its impact on cosmological inference.

Dey, Biprateep [University of Toronto; University ↗

Discovering faint and high apparent motion rate near-Earth asteroids using a deep learning program

ABSTRACT Although many near-Earth objects have been found by ground-based telescopes, some fast-moving ones, especially those near detection limits, have been missed by observatories. We developed a convolutional neural network for detecting faint fast-moving near-Earth objects. It was trained with artificial streaks generated from simulations and was able to find these asteroid streaks with an accuracy of 98.7 per cent and a false positive rate of 0.02 per cent on simulated data. This program was used to search image data from the Zwicky Transient Facility (ZTF) in four nights in 2019, and it identified six previously undiscovered asteroids. The visual magnitudes of our detections range from ∼19.0 to 20.3 and motion rates range from ∼6.8 to 24 deg d−1, which is very faint compared to other ZTF detections moving at similar motion rates. Our asteroids are also ∼1–51 m diameter in size and ∼5–60 lunar distances away at close approach, assuming their albedo values follow the albedo distribution function of known asteroids. The use of a purely simulated data set to train our model enables the program to gain sensitivity in detecting faint and fast-moving objects while still being able to recover nearly all discoveries made by previously designed neural networks which used real detections to train neural networks. Our approach can be adopted by any observatory for detecting fast-moving asteroid streaks.

Wang, Franklin (ORCID:0000000344887355)↗

DeepONet-grid-UQ: A trustworthy deep operator framework for predicting the power grid’s post-fault trajectories

This paper proposes a novel data-driven method for the reliable prediction of the power grid’s post-fault trajectories, i.e., the power grid’s dynamic response after a disturbance or fault. Here, the proposed method is based on the recently proposed concept of Deep Operator Networks (DeepONets). Unlike traditional neural networks that learn to approximate functions, DeepONets are designed to approximate nonlinear operators, i.e., mappings between infinite-dimensional spaces. Under this operator framework, we design a novel and efficient DeepONet that (i) takes as inputs the trajectories collected before and during the fault and (ii) outputs the predicted post-fault trajectories. In addition, we endow our method with the much-needed ability to balance efficiency with reliable/trustworthy predictions via uncertainty quantification. To this end, we propose and compare two novel methods that enable quantifying the predictive uncertainty. First, we propose a Bayesian DeepONet (B-DeepONet) that uses stochastic gradient Hamiltonian Monte-Carlo to sample from the posterior distribution of the DeepONet trainable parameters. Then, we design a Probabilistic DeepONet (Prob-DeepONet) that uses a probabilistic training strategy to enable quantifying uncertainty at virtually no extra computational cost. Finally, we validate the proposed methods’ predictive power and uncertainty quantification capability using the New York-New England power grid model.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Stakeholder Training for IEEE 1547-2018 (Abbreviated Final Technical Report)

This project identifies and makes publicly available educational materials to assist stakeholders in adopting the revised IEEE Std 1547-2018, Standard for Interconnection and Interoperability of Distributed Energy Resources with Associated Electric Power Systems Interfaces. The new standard is significantly different than the previous 2003 version, contains new concepts and new technical requirements, and requires educational material on the application of these changes. NREL had a principal role in accelerating the revision process though its chairmanship of the IEEE Std 1547 working group to revise the standard and through technical support provided to the working group. Under this project, NREL and partners compiled and developed publicly available educational materials for IEEE Std 1547-2018, including webinars, white papers, and other resources on related topics.

14 SOLAR ENERGY↗

Robust and scalable uncertainty estimation with conformal prediction for machine-learned interatomic potentials

Abstract Uncertainty quantification (UQ) is important to machine learning (ML) force fields to assess the level of confidence during prediction, as ML models are not inherently physical and can therefore yield catastrophically incorrect predictions. Established a-posteriori UQ methods, including ensemble methods, the dropout method, the delta method, and various heuristic distance metrics, have limitations such as being computationally challenging for large models due to model re-training. In addition, the uncertainty estimates are often not rigorously calibrated. In this work, we propose combining the distribution-free UQ method, known as conformal prediction (CP), with the distances in the neural network’s latent space to estimate the uncertainty of energies predicted by neural network force fields. We evaluate this method (CP+latent) along with other UQ methods on two essential aspects, calibration, and sharpness, and find this method to be both calibrated and sharp under the assumption of independent and identically-distributed (i.i.d.) data. We show that the method is relatively insensitive to hyperparameters selected, and test the limitations of the method when the i.i.d. assumption is violated. Finally, we demonstrate that this method can be readily applied to trained neural network force fields with traditional and graph neural network architectures to obtain estimates of uncertainty with low computational costs on a training dataset of 1 million images to showcase its scalability and portability. Incorporating the CP method with latent distances offers a calibrated, sharp and efficient strategy to estimate the uncertainty of neural network force fields. In addition, the CP approach can also function as a promising strategy for calibrating uncertainty estimated by other approaches.

74 ATOMIC AND MOLECULAR PHYSICS↗

Anomaly Detection, Localization and Classification using Drifting Synchrophasor Data Streams

With ongoing automation and digitization of the electric power system, several Phasor Measurement Units(PMUs) have been deployed for monitoring and control. PMU data can have multiple anomalies, and many of the researchers in the past have concentrated on training machine/deep learning algorithms offline for anomaly detection over PMU data (i.e., not in real time). These machine/deep learning algorithms, when trained offline on a sample rather than a population of the dataset, fail to consider the dynamic behavior of the power grid in real-time, resulting in low accuracy. Considering the dynamic behavior of the power grid (e.g., change in load, generation, distributed energy resources (DERs) switching, network, controls), the definition of data anomalies varies in time and requires online training. A fundamental challenge is to enable online (i.e., real-time) training of machine/deep learning algorithms for anomaly detection over streaming PMU data. While machine/deep learning is often desirable to manage data streams, training a deep learning algorithm over streaming PMU data is nontrivial due to changes in data statistics caused by dynamic streaming data. This paper proposes PMUNET: a novel device-level deep learning-based data-driven approach for anomaly detection, localization, and classification over streaming PMU data, using online learning and multivariate data-drift detection algorithm .Two variants of PMUNET, Dynamic data Change Driven Learning (DCDL) and Continuity Driven Learning (CDL), are proposed and compared. DCDL aims to train the deep learning algorithm whenever the definition of anomaly changes due to the power grid dynamics. On the other hand, CDL continuously trains the deep learning algorithm over the PMU data-stream. The experimental results verify that DCDL outperforms CDL and other efficient anomaly detection methods over multiple events such as faults and load/ generator/capacitor/DERs variations/switching for IEEE 14 and 39 Bus test system as well as real PMU industrial data. The result verifies that DCDL variant of PMUNET improves over existing approach with a gain of 2% - 10% in terms of accuracy, false-positive rate, and false-negative rate.

adversarial deep learning↗

Training custom light curve models of SN Ia subpopulations selected according to host galaxy properties

ABSTRACT Type Ia supernova (SN Ia) cosmology analyses include a luminosity step function in their distance standardization process to account for an observed yet unexplained difference in the post-standardization luminosities of SNe Ia originating from different host galaxy populations [e.g. high-mass ($M \gtrsim 10^{10} \, {\rm M}_{\odot }$) versus low-mass galaxies]. We present a novel method for including host-mass correlations in the SALT3 (Spectral Adaptive Light curve Template 3) light curve model used for standardizing SN Ia distances. We split the SALT3 training sample according to host-mass, training independent models for the low- and high-host-mass samples. Our models indicate that there are different average Si ii spectral feature strengths between the two populations, and that the average spectral energy distribution of SNe from low-mass galaxies is bluer than the high-mass counterpart. We then use our trained models to perform an SN cosmology analysis on the 3-yr spectroscopically confirmed Dark Energy Survey SN sample, treating SNe from low- and high-mass host galaxies as separate populations throughout. We find that our mass-split models reduce the Hubble residual scatter in the sample, albeit at a low statistical significance. We do find a reduction in the mass-correlated luminosity step but conclude that this arises from the model-dependent re-definition of the fiducial SN absolute magnitude rather than the models themselves. Our results stress the importance of adopting a standard definition of the SN parameters (x0, x1, c) in order to extract the most value out of the light curve modelling tools that are currently available and to correctly interpret results that are fit with different models.

Taylor, G. (ORCID:0000000157563259)↗

Evaluation of data driven low-rank matrix factorization for accelerated solutions of the Vlasov equation

Low-rank methods have shown success in accelerating simulations of a collisionless plasma described by the Vlasov equation, but still rely on computationally costly linear algebra every time step. We propose a data-driven factorization method using artificial neural networks, specifically with convolutional layer architecture, that trains on existing simulation data. At inference time, the model outputs a low-rank decomposition of the distribution field of the charged particles, and we demonstrate that this step is faster than the standard linear algebra technique. Numerical experiments show that the method achieves comparable reconstruction accuracy for interpolation tasks, generalizing to unseen test data in a manner beyond just memorizing training data; patterns in factorization also inherently followed the same numerical trend as those within algebraic methods (e.g., truncated singular-value decomposition). However, when training on the first 70% of a time-series data and testing on the remaining 30%, the method fails to meaningfully extrapolate. Despite this limiting result, the technique may have benefits for simulations in a statistical steady-state or otherwise showing temporal stability. These results suggest that while the model offers a computationally efficient alternative for datasets with temporal stability, its current formulation is best suited for interpolation rather than for predicting future states in time-evolving systems. This study thus lays the groundwork for further refinement of neural network-based approaches to low-rank matrix factorization in high-dimensional plasma simulations.

97 MATHEMATICS AND COMPUTING↗