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 253 records · Page 14

SAGIPS: A scalable Framework for scidac quantom

As part of the Scientific Discovery through Advanced Computing (SciDAC) program, the Quantum Chromodynamics Nuclear Tomography (QuantOM) project aims to analyze data from Deep Inelastic Scattering (DIS) experiments conducted at Thomas Jefferson National Accelerator Facility and the upcoming Electron Ion Collider. The DIS data analysis is performed on an event level by taking into leveraging nuclear theory models and accounting for experimental conditions. In order to efficiently run multiple analyses under varying conditions, a composable workflow was designed where each section (theory, experiment, objective minimization, etc.) has its own dedicated module. This presentation gives an overview over of the current status of this workflow, highlights present and future challenges, and highlights possible extensions to other projects with similar requirements.

Lersch, Daniel [Thomas Jefferson National Accelera↗

Digital Engineering & Data Acquisition

This is an Intern Poster submission regarding my research in the digital engineering team. My task was to create a software adapter to communicate between a data acquisition system (LabVIEW) and a data warehouse (Deep Lynx) as described in the poster.

99 GENERAL AND MISCELLANEOUS↗

A deep learning approach to real-time HIV outbreak detection using genetic data

Pathogen genomic sequence data are increasingly made available for epidemiological monitoring. A main interest is to identify and assess the potential of infectious disease outbreaks. While popular methods to analyze sequence data often involve phylogenetic tree inference, they are vulnerable to errors from recombination and impose a high computational cost, making it difficult to obtain real-time results when the number of sequences is in or above the thousands. Here, we propose an alternative strategy to outbreak detection using genomic data based on deep learning methods developed for image classification. The key idea is to use a pairwise genetic distance matrix calculated from viral sequences as an image, and develop convolutional neutral network (CNN) models to classify areas of the images that show signatures of active outbreak, leading to identification of subsets of sequences taken from an active outbreak. We showed that our method is efficient in finding HIV-1 outbreaks with R0 ≥ 2.5, and overall a specificity exceeding 98% and sensitivity better than 92%. We validated our approach using data from HIV-1 CRF01 in Europe, containing both endemic sequences and a well-known dual outbreak in intravenous drug users. Our model accurately identified known outbreak sequences in the background of slower spreading HIV. Importantly, we detected both outbreaks early on, before they were over, implying that had this method been applied in real-time as data became available, one would have been able to intervene and possibly prevent the extent of these outbreaks. This approach is scalable to processing hundreds of thousands of sequences, making it useful for current and future real-time epidemiological investigations, including public health monitoring using large databases and especially for rapid outbreak identification.

59 BASIC BIOLOGICAL SCIENCES↗

Deep Neural Networks are Adaptive to Function Regularity and Data Distribution in Approximation and Estimation

Deep learning has exhibited remarkable results across diverse areas. To understand its success, substantial research has been directed towards its theoretical foundations. Nev- ertheless, the majority of these studies examine how well deep neural networks can model functions with uniform regularities. In this paper, we explore a different angle: how deep neural networks can adapt to varying degrees of smoothness in functions and nonuni- form data distributions across different locations and scales. More precisely, we focus on a broad class of functions defined by nonlinear tree-based approximation methods. This class encompasses a range of function types, such as functions with uniform regularities and discontinuous functions. We develop nonparametric approximation and estimation theories for this class using deep ReLU networks. Our results show that deep neural networks are adaptive to the nonuniform smoothness of functions and nonuniform data distributions at different locations and scales. We apply our results to several function classes, and derive the corresponding approximation and generalization errors. The validity of our results is demonstrated through numerical experiments.

97 MATHEMATICS AND COMPUTING↗

The satellite population around luminous red galaxies in the 25 deg2 DESI Legacy Imaging Surveys Early Data Release

ABSTRACT Luminous red galaxies, or LRGs, are representative of the most massive galaxies and were originally selected in the Sloan Digital Sky Survey as good tracers of large-scale structure. They are dominated by by uniformly old stellar populations, have low star formation rates, early type morphologies, and little cold gas. Despite having old stellar populations and little in situ star formation, studies have shown that they have grown their stellar mass since z = 1, implying that they grow predominantly via the accretion of satellites. Tests of this picture have been limited because of the lack of deep imaging data sets that both covers a large enough area of the sky to contain substantial numbers of LRGs and that also is deep enough to detect faint satellites. We use the 25 deg2 Early Data Release (EDR) of the DESI Legacy Imaging Surveys to characterize the satellite galaxy population of LRGs out to z = 0.65. The DESI Legacy Imaging Surveys are comprised of grz imaging to 2–2.5 mag deeper than SDSS and with better image quality. We use a new statistical background technique to identify excess populations of putative satellite galaxies around 1823 LRGs at 0.2 < z < 0.65. In three redshift and luminosity bins we measure the numbers of satellite galaxies and their r−z colour distribution down to rest-frame g-band luminosity limits at least 3.6 times fainter than L*. In addition, we develop a forward modeling technique and apply it to constrain the mean number of satellites in each of our redshift and luminosity bins. Finally, we use these estimates to determine the amount of stellar mass growth in LRGs down to the local Universe.

Astronomy & Astrophysics↗

Discovery of Signatures, Anomalies, and Precursors in Synchrophasor Data with Matrix Profile and Deep Recurrent Neural Networks (Final Project Report)

The widespread deployment of phasor measurement unit (PMU) across the U.S. together with the burgeoning machine learning technology made it possible to develop data-driven PMU data analytics to improve grid security and reliability in a more insightful and effective manner. Although PMU applications have been explored for over a decade, the representative PMU usage is limited to the bulk power system monitoring mainly due to the data integrity issues associated with PMUs (typically missing, fragmented, and wrongly amplified data). To forge a breakthrough on this stalemate and embrace PMUs for power system control and protection as well, we applied various advanced machine learning and big data analysis technology to the power system event detection and classification as the first step toward the power system control and protection pertaining to grid security enhancement.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Dark Energy Survey Year 3 results: Simulation-based 𝑤CDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design

Data-driven approaches using deep learning are emerging as powerful techniques to extract non-Gaussian information from cosmological large-scale structure. Here, this work presents the first simulation-based inference (SBI) pipeline that combines weak lensing and galaxy clustering maps in a realistic Dark Energy Survey Year 3 (DES Y3) configuration and serves as preparation for a forthcoming analysis of the survey data. We develop a scalable forward model based on the CosmoGridV1 suite of N-body simulations to generate over one million self-consistent mock realizations of DES Y3 at the map level. Leveraging this large dataset, we train deep graph convolutional neural networks on the full survey footprint in spherical geometry to learn low-dimensional features that approximately maximize mutual information with target parameters. These learned compressions enable neural density estimation of the implicit likelihood via normalizing flows in a ten-dimensional parameter space spanning cosmological 𝑤CDM, intrinsic alignment, and linear galaxy bias parameters, while marginalizing over baryonic, photometric redshift, and shear bias nuisances. To ensure robustness, we extensively validate our inference pipeline using synthetic observations derived from both systematic contaminations in our forward model and independent Buzzard galaxy catalogs. Our forecasts yield significant improvements in cosmological parameter constraints, achieving 2−3× higher figures of merit in the 𝛺 𝑚 − 𝑆 8 plane relative to our implementation of baseline two-point statistics and effectively breaking parameter degeneracies through probe combination. These results demonstrate the potential of SBI analyses powered by deep learning for upcoming Stage-IV wide-field imaging surveys.

Thomsen, A. [Zurich, ETH] (ORCID:0000000203099021)↗

SUPERMASSIVE BLACK HOLE GROWTH IN STARBURST GALAXIES OVER COSMIC TIME: CONSTRAINTS FROM THE DEEPEST CHANDRA FIELDS

We present an analysis of deep multiwavelength data for z Almost-Equal-To 0.3-3 starburst galaxies selected by their 70 {mu}m emission in the Extended-Chandra Deep Field-South and Extended Groth Strip. We identify active galactic nuclei (AGNs) in these infrared sources through their X-ray emission and quantify the fraction that host an AGN. We find that the fraction depends strongly on both the mid-infrared color and rest-frame mid-infrared luminosity of the source, rising to {approx}50%-70% at the warmest colors (F{sub 24{mu}m}/F{sub 70{mu}m} {approx}< 0.2) and highest mid-infrared luminosities (corresponding to ultraluminous infrared galaxies), similar to the trends found locally. Additionally, we find that the AGN fraction depends strongly on the star formation rate (SFR) of the host galaxy (inferred from the observed-frame 70 {mu}m luminosity after subtracting the estimated AGN contribution), particularly for more luminous AGNs (L{sub 0.5{sub -8.0keV}} {approx}> 10{sup 43} erg s{sup -1}). At the highest SFRs ({approx}1000 M{sub Sun} yr{sup -1}), the fraction of galaxies with an X-ray detected AGN rises to Almost-Equal-To 30%, roughly consistent with that found in high-redshift submillimeter galaxies. Assuming that the AGN fraction is driven by the SFR (rather than stellar mass or redshift, for which our sample is largely degenerate), this result implies that the duty cycle of luminous AGN activity increases with the SFR of the host galaxy: specifically, we find that luminous X-ray detected AGNs are at least {approx}5-10 times more common in systems with high SFRs ({approx}> 300 M{sub Sun} yr{sup -1}) than in systems with lower SFRs ({approx}< 30 M{sub Sun} yr{sup -1}). Lastly, we investigate the ratio between the supermassive black hole accretion rate (inferred from the AGN X-ray luminosity) and the bulge growth rate of the host galaxy (approximated as the SFR) and find that, for sources with detected AGNs and star formation (and neglecting systems with low star formation rates to which our data are insensitive), this ratio in distant starbursts agrees well with that expected from the local scaling relation assuming the black holes and bulges grew at the same epoch. These results imply that black holes and bulges grow together during periods of vigorous star formation and AGN activity.

79 ASTRONOMY AND ASTROPHYSICS↗

Structural and Photometric Properties of the Andromeda Satellite Dwarf Galaxy Lacerta I from Deep Imaging with WIYN pODI

We present results from WIYN pODI imaging of Lacerta I (And XXXI), a satellite dwarf galaxy discovered in the outskirts of the Andromeda galaxy (M31) in Pan-STARRS1 survey data. Our deep, wide-field g,i photometry reaches ∼3 mag fainter than the photometry in the Pan-STARRS1 discovery paper and allows us to trace the stellar population of Lac I beyond two half-light radii from the galaxy center. We measure a Tip of the Red Giant Branch distance for Lac I of (m−M){sub 0}=24.44±0.11 mag (773 ± 40 kpc, or 264 ± 6 kpc from M31), which is consistent with the Pan-STARRS1 distance. We use a maximum-likelihood technique to derive structural properties for the galaxy, and find a half-light radius (r {sub h}) of 3.24 ± 0.21 arcmin (728 ± 47 pc), ellipticity (ϵ) of 0.44 ± 0.03, total magnitude M {sub V} = −11.4 ± 0.3, and central surface brightness μ{sub V,0}=24.8±0.3 mag arcsec{sup −2}. We find no H i emission in archival data and set a limit on Lac I’s neutral gas mass-to-light ratio of M{sub HI}/L{sub V} < 0.06 M{sub ⊙}/L{sub ⊙}, confirming Lac I as a gas-poor dwarf spheroidal galaxy. Photometric metallicities derived from Red Giant Branch stars within 2 r {sub h} yield a median [Fe/H] of −1.68 ± 0.03, which is more metal-rich than the spectroscopically derived value from Martin et al. Combining our measured magnitude with this higher metallicity estimate places Lac I closer to its expected position on the luminosity–metallicity relation for dwarf galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

A convolutional neural network-based screening tool for X-ray serial crystallography

A new tool is introduced for screening macromolecular X-ray crystallography diffraction images produced at an X-ray free-electron laser light source. Based on a data-driven deep learning approach, the proposed tool executes a convolutional neural network to detect Bragg spots. Automatic image processing algorithms described can enable the classification of large data sets, acquired under realistic conditions consisting of noisy data with experimental artifacts. Outcomes are compared for different data regimes, including samples from multiple instruments and differing amounts of training data for neural network optimization.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Automatic Calibration of a Geomechanical Model from Sparse Data for Estimating Stress in Deep Geological Formations

Summary In this study, we demonstrate geomechanical modeling with fully automatic parameter calibration to estimate the full geomechanical stress fields of a prospective US carbon dioxide (CO2) storage site, based on sparse measurement data. The goal is to compute full stress tensor field estimates (principal stresses and orientations) that are maximally compatible with observations within the constraints of the model assumptions, thereby extending pointwise, incomplete partial stress measurement to a simulated full formation stress field, as well as a rough assessment of the associated error. We use the Perch site, located in Otsego County, Michigan, USA, as our case study. The input data consist of partial stress tensor information inferred from in-situ borehole tests, geophysical well logs, and processing of seismic data. A static earth model (SEM) of the site was developed, and geomechanical simulation functionality of the open-source MATLAB Reservoir Simulation Toolbox (MRST) was used to model the stress field. Adjoint-based nonlinear optimization was used to adjust boundary conditions and material properties to calibrate simulated results of observations. Results were interpreted through a Bayesian framework. The focus of this paper is to demonstrate how the fully automatic calibration procedure works and discuss the results obtained; it does not attempt a detailed analysis of the stress field in the context of the proposed CO2 storage initiatives. Our work is part of a larger effort to noninvasively determine in-situ stresses in deep formations considered for CO2 storage. Guided by previously published research on geomechanical model calibration, our work presents a novel calibration approach supporting a potentially large number of linear or nonlinear calibration parameters to produce results optimally agreeing with available measurements and thus extend partial pointwise estimates to full tensor fields compatible with the physics of the site.

Engineering↗

A benchmark dataset for Hydrogen Combustion

The generation of reference data for deep learning models is challenging for reactive systems, and more so for combustion reactions due to the extreme conditions that create radical species and alternative spin states during the combustion process. Here, we extend intrinsic reaction coordinate (IRC) calculations with ab initio MD simulations and normal mode displacement calculations to more extensively cover the potential energy surface for 19 reaction channels for hydrogen combustion. A total of ~290,000 potential energies and ~1,270,000 nuclear force vectors are evaluated with a high quality range-separated hybrid density functional, ωB97X-V, to construct the reference data set, including transition state ensembles, for the deep learning models to study hydrogen combustion reaction.

08 HYDROGEN↗

Addressing GPU memory limitations for Graph Neural Networks in High-Energy Physics applications

Introduction Reconstructing low-level particle tracks in neutrino physics can address some of the most fundamental questions about the universe. However, processing petabytes of raw data using deep learning techniques poses a challenging problem in the field of High Energy Physics (HEP). In the Exa.TrkX Project, an illustrative HEP application, preprocessed simulation data is fed into a state-of-art Graph Neural Network (GNN) model, accelerated by GPUs. However, limited GPU memory often leads to Out-of-Memory (OOM) exceptions during training, due to the large size of models and datasets. This problem is exacerbated when deploying models on High-Performance Computing (HPC) systems designed for large-scale applications. Methods We observe a high workload imbalance issue during GNN model training caused by the irregular sizes of input graph samples in HEP datasets, contributing to OOM exceptions. We aim to scale GNNs on HPC systems, by prioritizing workload balance in graph inputs while maintaining model accuracy. Our paper introduces diverse balancing strategies aimed at decreasing the maximum GPU memory footprint and avoiding the OOM exception, across various datasets. Results Our experiments showcase memory reduction of up to 32.14% compared to the baseline. We also demonstrate the proposed strategies can avoid OOM in application. Additionally, we create a distributed multi-GPU implementation using these samplers to demonstrate the scalability of these techniques on the HEP dataset. Discussion By assessing the performance of these strategies as data loading samplers across multiple datasets, we can gauge their effectiveness in both single-GPU and distributed environments. Our experiments, conducted on datasets of varying sizes and across multiple GPUs, broaden the applicability of our work to various GNN applications that handle input datasets with irregular graph sizes.

Lee, Claire Songhyun↗

Deep Bayesian local crystallography

Abstract The advent of high-resolution electron and scanning probe microscopy imaging has opened the floodgates for acquiring atomically resolved images of bulk materials, 2D materials, and surfaces. This plethora of data contains an immense volume of information on materials structures, structural distortions, and physical functionalities. Harnessing this knowledge regarding local physical phenomena necessitates the development of the mathematical frameworks for extraction of relevant information. However, the analysis of atomically resolved images is often based on the adaptation of concepts from macroscopic physics, notably translational and point group symmetries and symmetry lowering phenomena. Here, we explore the bottom-up definition of structural units and symmetry in atomically resolved data using a Bayesian framework. We demonstrate the need for a Bayesian definition of symmetry using a simple toy model and demonstrate how this definition can be extended to the experimental data using deep learning networks in a Bayesian setting, namely rotationally invariant variational autoencoders.

36 MATERIALS SCIENCE↗

GMT: A deep learning approach to generalized multivariate translation for scientific data analysis and visualization

In scientific visualization, despite the significant advances of deep learning for data generation, researchers have not thoroughly investigated the issue of data translation. We present a new deep learning approach called generalized multivariate translation (GMT) for multivariate time-varying data analysis and visualization. Like V2V, GMT assumes a preprocessing step that selects suitable variables for translation. However, unlike V2V, which only handles one-to-one variable translation during training and inference, GMT enables one-to-many and many-to-many variable translation in the same framework. We leverage the recent StarGAN design from multi-domain image-to-image translation to achieve this generalization capability. We experiment with different loss functions and injection strategies to explore the best choices and leverage pre-training for performance improvement. We compare GMT with other state-of-the-art methods (i.e., Pix2Pix, V2V, StarGAN). Furthermore, the results demonstrate the overall advantage of GMT in translation quality and generalization ability.

97 MATHEMATICS AND COMPUTING↗

Machine Learning and Economic Models to Enable Risk-Informed Condition Based Maintenance of a Nuclear Plant Asset

The primary objective of this research is to address challenges in the implementation of risk-informed, condition-based predictive maintenance (PdM), which reduces operating costs while still maintaining the safety and reliability of commercial nuclear power plants (NPPs). To achieve the objective, risk models are being developed by taking advantage of advancements in data analytics, deep learning, machine learning (ML), and artificial intelligence (AI). The notable outcomes presented in the report include ? Development of a ML models using heterogeneous plant process and vibration data collected at different spatial and temporal resolutions from the Salem?s CWS to diagnose a circulating water pump (CWP) failure based on salient fault signatures. The developed diagnostic models are extendable to other faults associated with CWPs and CWP motors given associated fault signatures. ? Development of a natural language processing (NLP) technique to automatically classify the WO data into different categories. The developed NLP technique was validated on independent WO data. This automates the tedious and time-consuming activity of mining and classifying WOs by subject matter experts. ? Estimation of mean time between downtime (i.e., time duration between time instances when 1 or more CWPs are not available) and developed an approach to establish reliability of CWS components using unstructured WO data along with CWS plant process data. ? Formulation of economic model based on Markov chain models. The parameters of associated with the transition rate between different states of Markov chain models were estimated using WO data. The economic model formulation and discussion captures both time-independent and time-dependent parameter variation, leading to risk-informed decision-making.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗