Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Deep”

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 109 records · Page 6

DEEP KECK u-BAND IMAGING OF THE HUBBLE ULTRA DEEP FIELD: A CATALOG OF z approx 3 LYMAN BREAK GALAXIES

We present a sample of 407 z approx 3 Lyman break galaxies (LBGs) to a limiting isophotal u-band magnitude of 27.6 mag in the Hubble Ultra Deep Field. The LBGs are selected using a combination of photometric redshifts and the u-band drop-out technique enabled by the introduction of an extremely deep u-band image obtained with the Keck I telescope and the blue channel of the Low Resolution Imaging Spectrometer. The Keck u-band image, totaling 9 hr of integration time, has a 1sigma depth of 30.7 mag arcsec{sup -2}, making it one of the most sensitive u-band images ever obtained. The u-band image also substantially improves the accuracy of photometric redshift measurements of approx50% of the z approx 3 LBGs, significantly reducing the traditional degeneracy of colors between z approx 3 and z approx 0.2 galaxies. This sample provides the most sensitive, high-resolution multi-filter imaging of reliably identified z approx 3 LBGs for morphological studies of galaxy formation and evolution and the star formation efficiency of gas at high redshift.

79 ASTRONOMY AND ASTROPHYSICS↗

ULTRAVIOLET NUMBER COUNTS OF GALAXIES FROM SWIFT ULTRAVIOLET/OPTICAL TELESCOPE DEEP IMAGING OF THE CHANDRA DEEP FIELD SOUTH

Deep Swift UV/Optical Telescope (UVOT) imaging of the Chandra Deep Field South is used to measure galaxy number counts in three near-ultraviolet (NUV) filters (uvw2: 1928 A, uvm2: 2246 A, and uvw1: 2600 A) and the u band (3645 A). UVOT observations cover the break in the slope of the NUV number counts with greater precision than the number counts by the Hubble Space Telescope Space Telescope Imaging Spectrograph and the Galaxy Evolution Explorer, spanning a range 21 approx< m{sub AB} approx< 25. Model number counts confirm earlier investigations in favoring models with an evolving galaxy luminosity function.

79 ASTRONOMY AND ASTROPHYSICS↗

COLOR AND STELLAR POPULATION GRADIENTS IN PASSIVELY EVOLVING GALAXIES AT z {approx} 2 FROM HST/WFC3 DEEP IMAGING IN THE HUBBLE ULTRA DEEP FIELD

We report the detection of color gradients in six massive (stellar mass (M{sub star}) > 10{sup 10} M{sub sun}) and passively evolving (specific star formation rate <10{sup -11} yr{sup -1}) galaxies at redshift 1.3 < z < 2.5 identified in the Hubble Ultra Deep Field using ultra-deep Hubble Space Telescope (HST) Advanced Camera for Surveys and WFC3/IR images. After carefully matching the different point-spread functions, we obtain color maps and multi-band optical/near-IR photometry (BVizYJH) in concentric annuli, from the smallest resolved radial distance ({approx}1.7 kpc) up to several times the H-band effective radius. We find that the inner regions of these galaxies have redder rest-frame UV-optical colors (U - V, U - B, and B - V) than the outer parts. The slopes of the color gradient have no obvious dependence on the redshift and on the stellar mass of the galaxies. They do mildly depend, however, on the overall dust obscuration (E(B - V)) and rest-frame (U - V) color, with more obscured or redder galaxies having steeper color gradients. The z {approx} 2 color gradients are also steeper than those of local early-type ones. The gradient of a single parameter (age, extinction, or metallicity) cannot fully explain the observed color gradients. Fitting the spatially resolved HST seven-band photometry to stellar population synthesis models, we find that, regardless of assumptions on the metallicity gradient, the redder inner regions of the galaxies have slightly higher dust obscuration than the bluer outer regions, implying that dust partly contributes to the observed color gradients, although the magnitude depends on the assumed extinction law. Due to the age-metallicity degeneracy, the derived age gradient depends on the assumptions for the metallicity gradient. We discuss the implications of a number of assumptions for metallicity gradients on the formation and evolution of these galaxies. We find that the evolution of the mass-size relationship from z {approx} 2 to the present cannot be driven by in situ extended star formation, which implies that accretion or merger is mostly responsible for the growth of their stellar mass and size. The lack of a correlation between the strength of the color gradient and the stellar mass argues against the metallicity gradient predicted by the monolithic-collapse scenario, which would require significant major mergers to evolve into the one observed at the present.

79 ASTRONOMY AND ASTROPHYSICS↗

The Sloan Digital Sky Survey COADD: 275 deg{sup 2} of deep Sloan Digital Sky Survey imaging on stripe 82

We present details of the construction and characterization of the coaddition of the Sloan Digital Sky Survey (SDSS) Stripe 82 ugriz imaging data. This survey consists of 275 deg{sup 2} of repeated scanning by the SDSS camera over –50° ≤ α ≤ 60° and –1.°25 ≤ δ ≤ +1.°25 centered on the Celestial Equator. Each piece of sky has ∼20 runs contributing and thus reaches ∼2 mag fainter than the SDSS single pass data, i.e., to r ∼ 23.5 for galaxies. We discuss the image processing of the coaddition, the modeling of the point-spread function (PSF), the calibration, and the production of standard SDSS catalogs. The data have an r-band median seeing of 1.''1 and are calibrated to ≤1%. Star color-color, number counts, and PSF size versus modeled size plots show that the modeling of the PSF is good enough for precision five-band photometry. Structure in the PSF model versus magnitude plot indicates minor PSF modeling errors, leading to misclassification of stars as galaxies, as verified using VVDS spectroscopy. There are a variety of uses for this wide-angle deep imaging data, including galactic structure, photometric redshift computation, cluster finding and cross wavelength measurements, weak lensing cluster mass calibrations, and cosmic shear measurements.

79 ASTRONOMY AND ASTROPHYSICS↗

THE TAIWAN ECDFS NEAR-INFRARED SURVEY: ULTRA-DEEP J AND K{sub S} IMAGING IN THE EXTENDED CHANDRA DEEP FIELD-SOUTH

We present ultra-deep J and K{sub S} imaging observations covering a 30' Multiplication-Sign 30' area of the Extended Chandra Deep Field-South (ECDFS) carried out by our Taiwan ECDFS Near-Infrared Survey (TENIS). The median 5{sigma} limiting magnitudes for all detected objects in the ECDFS reach 24.5 and 23.9 mag (AB) for J and K{sub S} , respectively. In the inner 400 arcmin{sup 2} region where the sensitivity is more uniform, objects as faint as 25.6 and 25.0 mag are detected at 5{sigma}. Thus, this is by far the deepest J and K{sub S} data sets available for the ECDFS. To combine TENIS with the Spitzer IRAC data for obtaining better spectral energy distributions of high-redshift objects, we developed a novel deconvolution technique (IRACLEAN) to accurately estimate the IRAC fluxes. IRACLEAN can minimize the effect of blending in the IRAC images caused by the large point-spread functions and reduce the confusion noise. We applied IRACLEAN to the images from the Spitzer IRAC/MUSYC Public Legacy in the ECDFS survey (SIMPLE) and generated a J+K{sub S} -selected multi-wavelength catalog including the photometry of both the TENIS near-infrared and the SIMPLE IRAC data. We publicly release the data products derived from this work, including the J and K{sub S} images and the J+K{sub S} -selected multi-wavelength catalog.

79 ASTRONOMY AND ASTROPHYSICS↗

STRUCTURE AND MORPHOLOGIES OF z {approx} 7-8 GALAXIES FROM ULTRA-DEEP WFC3/IR IMAGING OF THE HUBBLE ULTRA-DEEP FIELD

We present a first morphological study of z {approx} 7-8 Lyman break galaxies (LBGs) from Oesch et al. and Bouwens et al. detected in ultra-deep near-infrared imaging of the Hubble Ultra-Deep Field (HUDF) by the HUDF09 program. With an average intrinsic size of 0.7 {+-} 0.3 kpc, these galaxies are found to be extremely compact, having an average observed surface brightness of {mu} {sub J} {approx_equal} 26 mag arcsec{sup -2}, and only two out of the full sample of 16 z {approx} 7 galaxies show extended features with resolved double cores. By comparison to lower redshift LBGs, it is found that only little size evolution takes place from z {approx} 7 to z {approx} 6, while galaxies between z {approx} 4-5 show more extended wings in their apparent profiles. The average size scales as (1 + z){sup -m} with m = 1.12 {+-} 0.17 for galaxies with luminosities in the range (0.3-1)L* {sub z=3} and with m = 1.32 {+-} 0.52 for (0.12-0.3)L* {sub z=3}, consistent with galaxies having constant comoving sizes. The peak of the size distribution changes only slowly from z {approx} 7 to z {approx} 4. However, a tail of larger galaxies ({approx}>1.2 kpc) is gradually built up toward later cosmic times, possibly via hierarchical build-up or via enhanced accretion of cold gas. Additionally, the average star formation surface density of LBGs with luminosities (0.3-1)L* {sub z=3} is nearly constant at {sigma}{sub SFR} = 1.9 M {sub sun} yr{sup -1} kpc{sup -2} over the entire redshift range z {approx} 4-7 suggesting similar star formation efficiencies at these early epochs. The above evolutionary trends seem to hold out to z {approx} 8 though the sample is still small and possibly incomplete.

79 ASTRONOMY AND ASTROPHYSICS↗

DISCOVERY OF z {approx} 8 GALAXIES IN THE HUBBLE ULTRA DEEP FIELD FROM ULTRA-DEEP WFC3/IR OBSERVATIONS

We utilize the newly acquired, ultra-deep WFC3/IR observations over the Hubble Ultra Deep Field (HUDF) to search for star-forming galaxies at z {approx} 8-8.5, only 600 million years from recombination, using a Y {sub 105}-dropout selection. The new 4.7 arcmin{sup 2} WFC3/IR observations reach to {approx}28.8 AB mag (5{sigma}) in the Y {sub 105} J {sub 125} H {sub 160} bands. These remarkable data reach {approx}1.5 AB mag deeper than the previous data over the HUDF, and now are an excellent match to the HUDF optical ACS data. For our search criteria, we use a two-color Lyman break selection technique to identify z {approx} 8-8.5Y {sub 105}-dropouts. We find five likely z {approx} 8-8.5 candidates. The sources have H {sub 160}-band magnitudes of {approx}28.3 AB mag and very blue UV-continuum slopes, with a median estimated {beta} of {approx}<-2.5 (where f {sub {lambda}} {proportional_to} {lambda}{sup {beta}}). This suggests that z {approx} 8 galaxies are not only essentially dust free but also may have very young ages or low metallicities. The observed number of Y {sub 105}-dropout candidates is smaller than the 20 {+-} 6 sources expected assuming no evolution from z {approx} 6, but is consistent with the five expected extrapolating the Bouwens et al. luminosity function (LF) results to z {approx} 8. These results provide evidence that the evolution in the LF seen from z {approx} 7 to z {approx} 3 continues to z {approx} 8. The remarkable improvement in the sensitivity of WFC3/IR has enabled Hubble Space Telescope to cross a threshold, revealing star-forming galaxies at z{approx} 8-9.

79 ASTRONOMY AND ASTROPHYSICS↗

DeepMerge – II. Building robust deep learning algorithms for merging galaxy identification across domains

In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations. Unfortunately, training a model on simulation data and then applying it to instrument data leads to a substantial and potentially even detrimental decrease in model accuracy on the new target dataset. Simulated and instrument data represent different data domains, and for an algorithm to work in both, domain-invariant learning is necessary. Here we employ domain adaptation techniques— Maximum Mean Discrepancy (MMD) as an additional transfer loss and Domain Adversarial Neural Networks (DANNs)— and demonstrate their viability to extract domain-invariant features within the astronomical context of classifying merging and non-merging galaxies. Additionally, we explore the use of Fisher loss and entropy minimization to enforce better in-domain class discriminability. We show that the addition of each domain adaptation technique improves the performance of a classifier when compared to conventional deep learning algorithms. We demonstrate this on two examples: between two Illustris-1 simulated datasets of distant merging galaxies, and between Illustris-1 simulated data of nearby merging galaxies and observed data from the Sloan Digital Sky Survey. The use of domain adaptation techniques in our experiments leads to an increase of target domain classification accuracy of up to ~20%. With further development, these techniques will allow astronomers to successfully implement neural network models trained on simulation data to efficiently detect and study astrophysical objects in current and future large-scale astronomical surveys.

galaxies: interactions↗

Resonance region structure functions and parity-violating deep inelastic scattering

The primary motive of parity-violating deep inelastic scattering experiments has been to test the standard model, particularly the axial couplings to the quarks, in the scaling region. The measurements can also test for the validity of models for the off-diagonal structure functions ${F}_{1,2,3}^{γZ}$($x,Q^2$) in the resonance region. The off-diagonal structure functions are important for the accurate calculation of the $γZ$-box correction to the weak charge of the proton. Currently, with no data to determine ${F}_{1,2,3}^{γZ}$($x,Q^2$) directly, models are constructed by modifying existing fits to electromagnetic data. Here, we present the asymmetry value for deuteron and proton targets predicted by several different ${F}_{1,2,3}^{γZ}$($x,Q^2$) models and demonstrate that there are notable disagreements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Deep-Learning Electron Diffractive Imaging

Here, we report the development of deep-learning coherent electron diffractive imaging at subangstrom resolution using convolutional neural networks (CNNs) trained with only simulated data. We experimentally demonstrate this method by applying the trained CNNs to recover the phase images from electron diffraction patterns of twisted hexagonal boron nitride, monolayer graphene, and a gold nanoparticle with comparable quality to those reconstructed by a conventional ptychographic algorithm. Fourier ring correlation between the CNN and ptychographic images indicates the achievement of a resolution in the range of 0.70 and 0.55 Å. We further develop CNNs to recover the probe function from the experimental data. The ability to replace iterative algorithms with CNNs and perform real-time atomic imaging from coherent diffraction patterns is expected to find applications in the physical and biological sciences.

47 OTHER INSTRUMENTATION↗

DeepHyper: Asynchronous Hyperparameter Search for Deep Neural Networks

Hyperparameters employed by deep learning (DL) methods play a substantial role in the performance and reliability of these methods in practice. Unfortunately, finding performance-optimizing hyperparameter settings is a notoriously difficult task. Hyperparameter search methods typically have limited production-strength implementations or do not target scalability within a highly parallel machine, portability across different machines, experimental comparison between different methods, and tighter integration with workflow systems. In this paper, we present DeepHyper, a Python package that provides a common interface for the implementation and study of scalable hyperparameter search methods. It adopts the Balsam workflow system to hide the complexities of running large numbers of hyperparameter configurations in parallel on high-performance computing (HPC) systems. We implement and study asynchronous model-based search methods that consist of sampling a small number of input hyperparameter configurations and progressively fitting surrogate models over the input-output space until exhausting a user-defined budget of evaluations. We evaluate the efficacy of these methods relative to approaches such as random search, genetic algorithms, Bayesian optimization, and hyperband on DL benchmarks on CPU- and GPU-based HPC systems.

Balaprakash, Prasanna↗

Deep Learning for Vertex Reconstruction of Neutrino-nucleus Interaction Events with Combined Energy and Time Data

We present a deep learning approach for vertex reconstruction of neutrino-nucleus interaction events, a problem in the domain of high energy physics. In this approach, we combine both energy and timing data that are collected in the MIN-ERvA detector to perform classification and regression tasks. We show that the resulting network achieves higher accuracy than previous results while requiring a smaller model size and less training time. In particular, the proposed model outperforms the state-of-the-art by 4.00% on classification accuracy. For the regression task, our model achieves 0.9919 on the coefficient of determination, higher than the previous work (0.96).

Song, Linghao↗

Scaling deep learning workloads: NVIDIA DGX-1/Pascal and Intel Knights Landing

Deep Learning (DL) algorithms have become ubiquitous in data analytics. As a result, major computing vendors --- including NVIDIA, Intel, AMD, and IBM --- have architectural road-maps influenced by DL workloads. Furthermore, several vendors have recently advertised new computing products as accelerating large DL workloads. Unfortunately, it is difficult for data scientists to quantify the potential of these different products. This paper provides a performance and power analysis of important DL workloads on two major parallel architectures: NVIDIA DGX-1 (eight Pascal P100 GPUs interconnected with NVLink) and Intel Knights Landing (KNL) CPUs interconnected with Intel Omni-Path or Cray Aries. Our evaluation consists of a cross section of convolutional neural net workloads: CifarNet, AlexNet, GoogLeNet, and ResNet50 topologies using the Cifar10 and ImageNet datasets. The workloads are vendor-optimized for each architecture. Our analysis indicates that although GPUs provide the highest overall performance, the gap can close for some convolutional networks; and the KNL can be competitive in performance/watt. We find that NVLink facilitates scaling efficiency on GPUs. However, its importance is heavily dependent on neural network architecture. Furthermore, for weak-scaling --- sometimes encouraged by restricted GPU memory --- NVLink is less important.

Gawande, Nitin A.↗

LeTID-Affected Cells from a Utility-Scale Photovoltaic System Characterized by Deep Level Transient Spectroscopy

Photovoltaic modules from a utility-scale field experienced power loss by light- and elevated temperature-induced degradation (LeTID). Samples of one of the affected monocrystalline silicon cells were cored and extracted from the module packaging and encapsulation. One of the cell fragments was processed using a regeneration cycle of applying short-circuit-rated current in forward bias at 85 degrees C for 2 weeks, while the other fragment was kept in its outdoor-degraded LeTID state. Both samples were scribed to form 2-mm diameter isolated areas using a femtosecond-pulse-width laser micromachining system. Both isolated areas contained front grid line segments which were wire bonded to larger contact pads, and the samples were probed in a cryostat linked to a deep-level transient spectroscopy (DLTS) system. Using DLTS, a majority-carrier, hole-trap defect was detected on each sample with an activation energy of 0.42 eV. The LeTID-degraded sample, however, had a larger signal corresponding to a trap density of 1.1x10^13 cm-3, which was about five times larger than the 2.1x10^12 cm-3 trap density of the regenerated sample.

charge carrier lifetime↗

A Novel Spatial-Temporal Variational Quantum Circuit to Enable Deep Learning on NISQ Devices

Quantum computing presents a promising approach for machine learning with its capability for extremely parallel computation in high-dimension through superposition and entanglement. Despite its potential, existing quantum learning algorithms, such as Variational Quantum Circuits (VQCs), face challenges in handling more complex datasets, particularly those that are not linearly separable. What’s more, it encounters the deployability issue, making the learning models suffer a drastic accuracy drop after deploying them to the actual quantum devices. To overcome these limitations, this paper proposes a novel spatial-temporal design, namely “ST-VQC”, to integrate nonlinearity in quantum learning and improve the robustness of the learning model to noise. Specifically, ST-VQC can extract spatial features via a novel block-based encoding quantum sub-circuit coupled with a layer-wise computation quantum sub-circuit to enable temporal-wise deep learning. Additionally, a SWAP-Free physical circuit design is devised to improve robustness. These designs bring a number of hyperparameters. After a systematic analysis of the design space for each design component, an automated optimization framework is proposed to generate the ST-VQC quantum circuit. The proposed ST-VQC has been evaluated on two IBM quantum processors, ibm-cairo with 27 qubits and ibmq-lima with 7 qubits to assess its effectiveness. The results of the evaluation on the standard dataset for binary classification show that ST-VQC can achieve over 30% accuracy improvement compared with existing VQCs on actual quantum computers. Moreover, on a non-linear synthetic dataset, the STVQC outperforms a linear classifier by 27.9%, while the linear classifier using classical computing outperforms the existing VQC by 15.58%.

Li, Jinyang↗

A laboratory nanoseismological study on deep-focus earthquake micromechanics

Global earthquake occurring rate displays an exponential decay down to ~300 km and then peaks around 550 to 600 km before terminating abruptly near 700 km. How fractures initiate, nucleate, and propagate at these depths remains one of the greatest puzzles in earth science, as increasing pressure inhibits fracture propagation. We report nanoseismological analysis on high-resolution acoustic emission (AE) records obtained during ruptures triggered by partial transformation from olivine to spinel in Mg 2 GeO 4 , an analog to the dominant mineral (Mg,Fe) 2 SiO 4 olivine in the upper mantle, using state-of-the-art seismological techniques, in the laboratory. AEs’ focal mechanisms, as well as their distribution in both space and time during deformation, are carefully analyzed. Microstructure analysis shows that AEs are produced by the dynamic propagation of shear bands consisting of nanograined spinel. These nanoshear bands have a near constant thickness (~100 nm) but varying lengths and self-organize during deformation. This precursory seismic process leads to ultimate macroscopic failure of the samples. Several source parameters of AE events were extracted from the recorded waveforms, allowing close tracking of event initiation, clustering, and propagation throughout the deformation/transformation process. AEs follow the Gutenberg-Richter statistics with a well-defined b value of 1.5 over three orders of moment magnitudes, suggesting that laboratory failure processes are self-affine. The seismic relation between magnitude and rupture area correctly predicts AE magnitude at millimeter scales. A rupture propagation model based on strain localization theory is proposed. Future numerical analyses may help resolve scaling issues between laboratory AE events and deep-focus earthquakes.

58 GEOSCIENCES↗

Dependence of Deep Convective Cell Properties on Meteorological and Aerosol Conditions during TRACER

Deep convective cells significantly influence Earth’s energy balance and water cycle. However, their accurate representation in numerical models remains challenging due to their small spatiotemporal scales and limited observational constraints. This study examines over ∼400 deep convective cells near Houston, observed by a dual-polarization C-band radar during the Tracking Aerosol Convection Interactions Experiment (TRACER) intensive observation period (June–September 2022). Cells are categorized by lifetime into short-lived (<40 min), intermediate-lived (40–80 min), and long-lived (80+ min) groups. Long-lived cells were broader (∼13.2 km at 2–4-km height) and deeper (∼11.4 km) than short-lived cells (∼6.4-km width, ∼7.31-km height). Using random forest (RF) modeling and correlation analyses, precipitable water vapor (PWV), 2–6-km lapse rate, 0–8-km bulk shear, and fine aerosol mass concentration (Mass_f) are identified as key predictors of cell lifetime. Higher PWV is associated with significantly longer convective cell lifetimes compared to the low-PWV group, particularly within low 2–6-km temperature lapse rate (LR_26km), moderate-to-higher 0–8-km bulk shear (BS_08km), and low-to-moderate Mass_f environments. RF analysis also identifies low-level (0–2 km) equivalent potential temperature, PWV, Mass_f, and surface latent heat flux as key predictors for cell width and height. Short-lived cells have higher aerosol number concentrations (500–1000-nm size range), linked to onshore wind conditions and marine aerosols; however, their low concentration suggests the sensitivity may reflect associated meteorological regimes rather than a direct aerosol effect. Long-lived cells have higher concentrations of organic and sulfate aerosols, while short-lived cells exhibit higher black carbon concentrations. These results highlight the intricate dependence of convective cell lifetimes and structure on environmental moisture, thermodynamics, wind shear, and aerosol characteristics.

54 ENVIRONMENTAL SCIENCES↗