Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep data”

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 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Accelerating Lossy and Lossless Compression on Emerging BlueField DPU Architectures

Data compression has become a crucial technique in addressing performance bottlenecks caused by increasing data volumes in High-Performance Computing (HPC), Big Data, and Deep Learning (DL). Despite its potential to boost system performance, recent studies have identified significant challenges with existing compression methods, mainly due to their high computational demands amidst continuously growing data sizes. Concurrently, the advent of Data Processing Units (DPUs), equipped with programmable System-on-Chip (SoC) and specialized compression accelerators, offers a promising opportunity to alter the landscape of data compression. This paper explores the complexities and potential of leveraging NVIDIA BlueField DPUs to accelerate lossy and lossless compression. Towards this, we introduce PEDAL, an innovative library that leverages the hardware capabilities of DPUs to unify and optimize data compression designs. Moreover, we seamlessly co-design PEDAL with the popular MPICH MPI library, demonstrating up to 101x speedup in compression time and 88x decrease in communication latency. Drawing on these achievements, we share our experience with various research communities about accelerating data compression on DPUs in communication-oriented HPC scenarios.

Li, Yuke↗
NASA NTRS2002

Emission Line Galaxies in the STIS Parallel Survey II: Star Formation Density

We present the luminosity function of [OII]-emitting galaxies at a median redshift of z = 0.9, as measured in the deep spectroscopic data in the STIS Parallel Survey (SPS). The luminosity function shows strong evolution from the local value, as expected. By using random lines of sight, the SPS measurement complements previous deep single field studies. We calculate the density of inferred star formation at this redshift by converting from [OII] to H(alpha) line flux as a function of absolute magnitude and find rho = 0.052 +/- 0.017 Solar mass/yr Mpc(sup -3) at a median redshift z approx. 0.9 within the range 0.46 less than z less than 1.415 (H(sub 0) = 50 km/s Mpc(sup -l), Omega(sub M) = 1.0, Omega(sub lambda) = 0.0). This density is consistent with a (1 + z)(sup )4 evolution in global star formation since z approx. 1. To reconcile the density with similar measurements made by surveys targeting H(alpha) may require substantial extinction correction.

Teplitz, Harry I.↗
DOE OSTI2019

The need for uncertainty quantification in machine-assisted medical decision making

We present that medicine, even from the earliest days of artificial intelligence (AI) research, has been one of the most inspiring and promising domains for the application of AI-based approaches. Equally, it has been one of the more challenging areas to see an effective adoption. There are many reasons for this, primarily the reluctance to delegate decision making to machine intelligence in cases where patient safety is at stake. To address some of these challenges, medical AI, especially in its modern data-rich deep learning guise, needs to develop a principled and formal uncertainty quantification (UQ) discipline, just as we have seen in fields such as nuclear stockpile stewardship and risk management. The data-rich world of AI-based learning and the frequent absence of a well-understood underlying theory poses its own unique challenges to straightforward adoption of UQ. These challenges, while not trivial, also present significant new research opportunities for the development of new theoretical approaches, and for the practical applications of UQ in the area of machine-assisted medical decision making. Finally, understanding prediction system structure and defensibly quantifying uncertainty is possible, and, if done, can significantly benefit both research and practical applications of AI in this critical domain.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

DL-TODA: A Deep Learning Tool for Omics Data Analysis

Metagenomics is a technique for genome-wide profiling of microbiomes; this technique generates billions of DNA sequences called reads. Given the multiplication of metagenomic projects, computational tools are necessary to enable the efficient and accurate classification of metagenomic reads without needing to construct a reference database. The program DL-TODA presented here aims to classify metagenomic reads using a deep learning model trained on over 3000 bacterial species. A convolutional neural network architecture originally designed for computer vision was applied for the modeling of species-specific features. Using synthetic testing data simulated with 2454 genomes from 639 species, DL-TODA was shown to classify nearly 75% of the reads with high confidence. The classification accuracy of DL-TODA was over 0.98 at taxonomic ranks above the genus level, making it comparable with Kraken2 and Centrifuge, two state-of-the-art taxonomic classification tools. DL-TODA also achieved an accuracy of 0.97 at the species level, which is higher than 0.93 by Kraken2 and 0.85 by Centrifuge on the same test set. Application of DL-TODA to the human oral and cropland soil metagenomes further demonstrated its use in analyzing microbiomes from diverse environments. Compared to Centrifuge and Kraken2, DL-TODA predicted distinct relative abundance rankings and is less biased toward a single taxon.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Transmission Line Fault Location Using Deep Learning Techniques

This letter presents a novel single-ended fault location approach for transmission lines that uses modern deep learning methods. A mixed convolutional neural network and long short-term memory structure are proposed to preserve the translation invariance and capture the temporal correlation of the time-series input data. Advanced deep learning techniques such as adaptive moment estimation and dropout are used to efficiently train the neural network and prevent over-fitting. Our extensive studies have demonstrated the accuracy and effectiveness of the deep-learning-based, singled-ended fault location approach.

Fan, Rui↗
DOE OSTI2022

The Data Synergy Effects of Time-Series Deep Learning Models in Hydrology

When fitting statistical models to variables in geoscientific disciplines such as hydrology, it is a customary practice to stratify a large domain into multiple regions (or regimes) and study each region separately. Traditional wisdom suggests that models built for each region separately will have higher performance because of homogeneity within each region. However, each stratified model has access to fewer and less diverse data points. Here, through two hydrologic examples (soil moisture and streamflow), we show that conventional wisdom may no longer hold in the era of big data and deep learning (DL). We systematically examined an effect we call data synergy, where the results of the DL models improved when data were pooled together from characteristically different regions. The performance of the DL models benefited from modest diversity in the training data compared to a homogeneous training set, even with similar data quantity. Moreover, allowing heterogeneous training data makes eligible much larger training datasets, which is an inherent advantage of DL. A large, diverse data set is advantageous in terms of representing extreme events and future scenarios, which has strong implications for climate change impact assessment. The results here suggest the research community should place greater emphasis on data sharing.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2016

AN EMPIRICAL DETERMINATION OF THE INTERGALACTIC BACKGROUND LIGHT FROM UV TO FIR WAVELENGTHS USING FIR DEEP GALAXY SURVEYS AND THE GAMMA-RAY OPACITY OF THE UNIVERSE

We have previously calculated the intergalactic background light (IBL) as a function of redshift from the Lyman limit in the far-ultraviolet to a wavelength of 5 μ m in the near-infrared range, based purely on data from deep galaxy surveys. Here, we use similar methods to determine the mid- and far-infrared IBL from 5 to 850 μ m. Our approach enables us to constrain the range of photon densities by determining the uncertainties in observationally determined luminosity densities and spectral gradients. By also including the effect of the 2.7 K cosmic background photons, we determine upper and lower limits on the opacity of the universe to γ -rays up to PeV energies within a 68% confidence band. Our direct results on the IBL are consistent with those from complimentary γ -ray analyses using observations from the Fermi γ -ray space telescope and the H.E.S.S. air Čerenkov telescope. Thus, we find no evidence of previously suggested processes for the modification of γ -ray spectra other than that of absorption by pair production alone.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Pan-cancer proteogenomics characterization of tumor immunity

Despite the successes of immunotherapy in cancer treatment over recent decades, less than <10%–20% cancer cases have demonstrated durable responses from immune checkpoint blockade. To enhance the efficacy of immunotherapies, combination therapies suppressing multiple immune evasion mechanisms are increasingly contemplated. To better understand immune cell surveillance and diverse immune evasion responses in tumor tissues, we comprehensively characterized the immune landscape of more than 1,000 tumors across ten different cancers using CPTAC pan-cancer proteogenomic data. We identified seven distinct immune subtypes based on integrative learning of cell type compositions and pathway activities. We then thoroughly categorized unique genomic, epigenetic, transcriptomic, and proteomic changes associated with each subtype. Further leveraging the deep phosphoproteomic data, we studied kinase activities in different immune subtypes, which revealed potential subtype-specific therapeutic targets. Insights from this work will facilitate the development of future immunotherapy strategies and enhance precision targeting with existing agents.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2017

White Paper on Nuclear Data Needs and Capabilities for Basic Science

Reliable nuclear structure and reaction data represent the fundamental building blocks of nuclear physics and astrophysics research, and are also of importance in many applications. There is a continuous demand for high-quality updates of the main nuclear physics databases via the prompt compilation and evaluation of the latest experimental and theoretical results. The nuclear physics research community benefits greatly from comprehensive, systematic and up-to-date reviews of the experimentally determined nuclear properties and observables, as well as from the ability to rapidly access these data in user-friendly forms. Such credible databases also act as a bridge between science, technology, and society by making the results of basic nuclear physics research available to a broad audience of users, and hence expand the societal utilization of nuclear science. Compilation and evaluation of nuclear data has deep roots in the history of nuclear science research, as outlined in Appendix 1. They have an enormous impact on many areas of science and applications, as illustrated in Figure 2 for the Evaluated Nuclear Structure Data File (ENSDF) database. The present workshop concentrated on the needs of the basic nuclear science community for data and capabilities. The main role of this community is to generate and use data in order to understand the basic nuclear forces and interactions that are responsible for the existence and the properties of all nuclides and, as a consequence, to gain knowledge about the origins, evolution and structure of the universe. Thus, the experiments designed to measure a wealth of nuclear properties towards these fundamental scientific goals are typically performed from within this community.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

A Data-Driven Framework for Power System Event Type Identification via Safe Semi-Supervised Techniques

Herein this paper investigates the use of phasor measurement unit (PMU) data with deep learning techniques to construct real-time event identification models for transmission networks. Increasing penetration of distributed energy resources represents a great opportunity to achieve decarbonization, as well as challenges in systematic situational awareness. When high-resolution PMU data and sufficient manually recorded event labels are available, the power event identification problem is defined as a statistical classification problem that can be solved by numerous cutting-edge classifiers. However, in real grids, collecting tremendous high-quality event labels is quite expensive. Utilities frequently have a large number of event records without in-depth details (i.e., unlabeled events). To bridge this gap, we propose a novel semi-supervised learning-based method to improve the performance of event classifiers trained with a limited number of labeled events by exploiting the information from massive unlabeled events. In other words, compared to existing data-driven methods, our method requires only a small portion of labeled data to achieve a similar level of accuracy. Meanwhile, this work discusses and addresses the performance degradation caused by class distribution mismatch between the training set and the real applications. Based on the proposed safe learning mechanism, our model does not directly use all unlabeled events during model training, but selectively uses them through a comprehensive evaluation procedure. Numerical studies on a sizable PMU dataset have been used to validate the performance of the proposed method.

42 ENGINEERING↗
DOE OSTI2018

Extraction of partonic transverse momentum distributions from semi-inclusive deep inelastic scattering and Drell-Yan data

We present a first attempt at a global fit of unpolarized quark transverse momentum dependent distribution and fragmentation functions from available data on semi-inclusive deep-inelastic scattering, Drell-Yan and $Z$ boson production processes. This analysis is performed in the low transverse momentum region, at leading order in perturbative QCD and with the inclusion of energy scale evolution effects at the next-to-leading logarithmic accuracy.

Pisano, Cristian↗
NASA NTRS1976

Seasat-A Synthetic Aperture Radar - Radar system implementation

The Synthetic Aperture Radar (SAR) onboard the Seasat-A satellite will conduct a number of experiments involving deep ocean waves, coastal wave patterns, polar ice and land forms. The SAR will have a 25 m by 25 m resolution over a swath of 100 km width centered about 300 km to the right of the spacecraft track. The SAR's high data rate limits operations to times when Seasat-A is in view of a few ground stations with special SAR receiving equipment. However, the SAR will collect much useful data about deep ocean and coastal waves in the Atlantic and Pacific Oceans; about ice in the Northwest Atlantic, in the Great Lakes and off the coast of Alaska; and about land over much of the United States and Canada.

Thompson, T. W.↗
NASA NTRS2011

Aircraft Lightning Electromagnetic Environment Measurement

This paper outlines a NASA project plan for demonstrating a prototype lightning strike measurement system that is suitable for installation onto research aircraft that already operate in thunderstorms. This work builds upon past data from the NASA F106, FAA CV-580, and Transall C-180 flight projects, SAE ARP5412, and the European ILDAS Program. The primary focus is to capture airframe current waveforms during attachment, but may also consider pre and post-attachment current, electric field, and radiated field phenomena. New sensor technologies are being developed for this system, including a fiber-optic Faraday polarization sensor that measures lightning current waveforms from DC to over several Megahertz, and has dynamic range covering hundreds-of-volts to tens-of-thousands-of-volts. A study of the electromagnetic emission spectrum of lightning (including radio wave, microwave, optical, X-Rays and Gamma-Rays), and a compilation of aircraft transfer-function data (including composite aircraft) are included, to aid in the development of other new lightning environment sensors, their placement on-board research aircraft, and triggering of the onboard instrumentation system. The instrumentation system will leverage recent advances in high-speed, high dynamic range, deep memory data acquisition equipment, and fiber-optic interconnect.

Ely, Jay J.↗
NASA NTRS2015

GeneLab: A Systems Biology Platform for Spaceflight Omics Data

NASA's mission includes expanding our understanding of biological systems to improve life on Earth and to enable long-duration human exploration of space. Resources to support large numbers of spaceflight investigations are limited. NASA's GeneLab project is maximizing the science output from these experiments by: (1) developing a unique public bioinformatics database that includes space bioscience relevant "omics" data (genomics, transcriptomics, proteomics, and metabolomics) and experimental metadata; (2) partnering with NASA-funded flight experiments through bio-sample sharing or sample augmentation to expedite omics data input to the GeneLab database; and (3) developing community-driven reference flight experiments. The first database, GeneLab Data System Version 1.0, went online in April 2015. V1.0 contains numerous flight datasets and has search and download capabilities. Version 2.0 will be released in 2016 and will link to analytic tools. In 2015 Genelab partnered with two Biological Research in Canisters experiments (BBRIC-19 and BRIC-20) which examine responses of Arabidopsis thaliana to spaceflight. GeneLab also partnered with Rodent Research-1 (RR1), the maiden flight to test the newly developed rodent habitat. GeneLab developed protocols for maxiumum yield of RNA, DNA and protein from precious RR-1 tissues harvested and preserved during the SpaceX-4 mission, as well as from tissues from mice that were frozen intact during spaceflight and later dissected. GeneLab is establishing partnerships with at least three planned flights for 2016. Organism-specific nationwide Science Definition Teams (SDTs) will define future GeneLab dedicated missions and ensure the broader scientific impact of the GeneLab missions. GeneLab ensures prompt release and open access to all high-throughput omics data from spaceflight and ground-based simulations of microgravity and radiation. Overall, GeneLab will facilitate the generation and query of parallel multi-omics data, and deep curation of metadata for integrative analysis, allowing researchers to uncover cellular networks as observed in systems biology platforms. Consequently, the scientific community will have access to a more complete picture of functional and regulatory networks responsive to the spaceflight environment.. Analysis of GeneLab data will contribute fundamental knowledge of how the space environment affects biological systems, and enable emerging terrestrial benefits resulting from mitigation strategies to prevent effects observed during exposure to space. As a result, open access to the data will foster new hypothesis-driven research for future spaceflight studies spanning basic science to translational science.

proteomics↗
DOE OSTI2024

Characterizing climate pathways using feature importance on echo state networks

The 2022 National Defense Strategy of the United States listed climate change as a serious threat to national security. Climate intervention methods, such as stratospheric aerosol injection, have been proposed as mitigation strategies, but the downstream effects of such actions on a complex climate system are not well understood. The development of algorithmic techniques for quantifying relationships between source and impact variables related to a climate event (i.e., a climate pathway) would help inform policy decisions. Data-driven deep learning models have become powerful tools for modeling highly nonlinear relationships and may provide a route to characterize climate variable relationships. In this paper, we explore the use of an echo state network (ESN) for characterizing climate pathways. ESNs are a computationally efficient neural network variation designed for temporal data, and recent work proposes ESNs as a useful tool for forecasting spatiotemporal climate data. However, ESNs are noninterpretable black-box models along with other neural networks. The lack of model transparency poses a hurdle for understanding variable relationships. We address this issue by developing feature importance methods for ESNs in the context of spatiotemporal data to quantify variable relationships captured by the model. We conduct a simulation study to assess and compare the feature importance techniques, and we demonstrate the approach on reanalysis climate data. In the climate application, we consider a time period that includes the 1991 volcanic eruption of Mount Pinatubo. This event was a significant stratospheric aerosol injection, which acts as a proxy for an anthropogenic stratospheric aerosol injection. Furthermore, we are able to use the proposed approach to characterize relationships between pathway variables associated with this event that agree with relationships previously identified by climate scientists.

black-box models↗
DOE OSTI2018

Utah FORGE: Deep Well 58-32 (MU-ESW1) Core Data

These datasets, images, and graphics were derived from core drilling and core that was extracted from Utah Forge deep well 58-32 (originally called MU-ESW1), near Roosevelt Hot Springs.

15 GEOTHERMAL ENERGY↗
NASA NTRS2011

Stardust-NExT, Deep Impact, and the Accelerating Spin of 9P/Tempel One

The evolution of the spin rate of comet 9P/Tempel 1 through two perihelion passages (CYs 2000 and 2005) is determined from 1922 Earth-based observations taken over a period of 13y as part of a World-Wide observing campaign and 2888 observations taken over a period of 50d from the Deep Impact spacecraft. We determine the following sidereal spin rates (periods): 209.023 +/- 0.025 degrees /day (41.335 +/- 0.005 h) prior to the 2000 perihelion passage, 210.448 +/- 0.016 degrees/day (41.055 +/- 0.003 h) for the interval between the 2000 and 2005 perihelion passages, 211.856 +/- 0.030 degrees/day (40.783 +/- 0.006 h) from Deep Impact photometry just prior to the 2005 perihelion passage, and 211.625 +/- 0.012 degrees /day (40.827 +/- 0.002 h) in the interval 2006-2010 following the 2005 perihelion passage. The period decreased by 16.8 +/- 0.3 min during the 2000 passage and by 13.7 +/- 0.2 min during the 2005 passage suggesting a secular decrease in the net torque. The change in spin rate is asymmetric with respect to perihelion with the maximum net torque being applied on approach to perihelion. The Deep Impact data alone show that the spin rate was increasing at a rate of 0.024 +/- 0.003 degree/d/d at JD2453530.60510 (i.e., 25.134 d before impact) and provides independent confirmation of the change seen in the Earth-based observations. The rotational phase of the nucleus at times before and after each perihelion and at the Deep Impact encounter is estimated based on the Thomas et al. pole and longitude system. The possibility of a 180 degree error in the rotational phase is assessed and found to be significant. Analytical and physical modeling of the behavior of the spin rate through of each perihelion is presented and used as a basis to predict the rotational state of the nucleus at the time of the nominal (i.e., prior to February 2010) Stardust-NExT encounter on 2011 February 14 20:42. We find that a net torque in the range of 0.3 - 2.5 x 10(exp 7) kg.square m2/square s acts on the nucleus during perihelion passage. The spin rate initially slows down on approach to perihelion and then passes through a minimum. It then accelerates rapidly as it passes through perihelion eventually reaching a maximum post-perihelion. It then decreases to a stable value as the nucleus moves away from the sun. We find that the pole direction is unlikely to precess by more than approximately 1 degree/perihelion passage. The trend of the period with time and the fact that the modeled peak torque that occurs before perihelion is in agreement with published accounts of trends in water production rate and suggests that widespread H2O out-gassing from the surface is largely responsible for the observed spin-up.

Belton, Michael J. S.↗
NASA NTRS2011

Stardust-NExT, Deep Impact, and the Accelerating Spin of 9P/Tempel 1

The evolution of the spin rate of Comet 9P/Tempel 1 through two perihelion passages (in 2000 and 2005) is determined from 1922 Earth-based observations taken over a period of 13 year as part of a World-Wide observing campaign and from 2888 observations taken over a period of 50 days from the Deep Impact spacecraft. We determine the following sidereal spin rates (periods): 209.023 +/- 0.025deg/dy (41.335 0.005 h) prior to the 2000 perihelion passage, 210.448 +/- 0.016deg/dy (41.055 +/- 0.003 h) for the interval between the 2000 and 2005 perihelion passages, 211.856 +/- 0.030deg/dy (40.783 +/- 0.006 h) from Deep Impact photometry just prior to the 2005 perihelion passage, and 211.625 +/- 0.012deg/dy (40.827 +/- 0.002 h) in the interval 2006-2010 following the 2005 perihelion passage. The period decreased by 16.8 +/- 0.3 min during the 2000 passage and by 13.7 +/- 0.2 min during the 2005 passage suggesting a secular decrease in the net torque. The change in spin rate is asymmetric with respect to perihelion with the maximum net torque being applied on approach to perihelion. The Deep Impact data alone show that the spin rate was increasing at a rate of 0.024 +/- 0.003deg/dy/dy at JD2453530.60510 (i.e., 25.134 dy before impact), which provides independent confirmation of the change seen in the Earth-based observations. The rotational phase of the nucleus at times before and after each perihelion and at the Deep Impact encounter is estimated based on the Thomas et al. (Thomas et al. [2007]. Icarus 187, 4-15) pole and longitude system. The possibility of a 180deg error in the rotational phase is assessed and found to be significant. Analytical and physical modeling of the behavior of the spin rate through of each perihelion is presented and used as a basis to predict the rotational state of the nucleus at the time of the nominal (i.e., prior to February 2010) Stardust-NExT encounter on 2011 February 14 at 20:42. We find that a net torque in the range of 0.3-2.5 x 10(exp 7) kg/sq m/sq s acts on the nucleus during perihelion passage. The spin rate initially slows down on approach to perihelion and then passes through a minimum. It then accelerates rapidly as it passes through perihelion eventually reaching a maximum post-perihelion. It then decreases to a stable value as the nucleus moves away from the Sun. We find that the pole direction is unlikely to precess by more than approx. 1deg per perihelion passage. The trend of the period with time and the fact that the modeled peak torque occurs before perihelion are in agreement with published accounts of trends in water production rate and suggests that widespread H2O out-gassing from the surface is largely responsible for the observed spin-up.

Belton, Michael J. S.↗