Search NASA⌕ Search

SEARCH · Search NASA

Results for “Tensor factorizations”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2011

SEVEN-YEAR WILKINSON MICROWAVE ANISOTROPY PROBE (WMAP ) OBSERVATIONS: POWER SPECTRA AND WMAP-DERIVED PARAMETERS

The WMAP mission has produced sky maps from seven years of observations at L2. We present the angular power spectra derived from the seven-year maps and discuss the cosmological conclusions that can be inferred from WMAP data alone. With the seven-year data, the temperature (TT) spectrum measurement has a signal-to-noise ratio per multipole that exceeds unity for l < 919; and in band powers of width {Delta}l = 10, the signal-to-noise ratio exceeds unity up to l = 1060. The third acoustic peak in the TT spectrum is now well measured by WMAP. In the context of a flat {Lambda}CDM model, this improvement allows us to place tighter constraints on the matter density from WMAP data alone, {Omega}{sub m} h {sup 2} = 0.1334{sup +0.0056}{sub -0.0055}, and on the epoch of matter-radiation equality, z{sub eq} = 3196{sup +134}{sub -133}. The temperature-polarization (TE) spectrum is detected in the seven-year data with a significance of 20{sigma}, compared to 13{sigma} with the five-year data. We now detect the second dip in the TE spectrum near l {approx} 450 with high confidence. The TB and EB spectra remain consistent with zero, thus demonstrating low systematic errors and foreground residuals in the data. The low-l EE spectrum, a measure of the optical depth due to reionization, is detected at 5.5{sigma} significance when averaged over l = 2-7: l(l + 1)C {sup EE}{sub l}/(2{pi}) = 0.074{sup +0.034}{sub -0.025} {mu}K{sup 2} (68% CL). We now detect the high-l, 24 {<=} l {<=} 800, EE spectrum at over 8{sigma}. The BB spectrum, an important probe of gravitational waves from inflation, remains consistent with zero; when averaged over l = 2-7, l(l + 1)C {sup BB}{sub l}/(2{pi}) < 0.055 {mu}K{sup 2} (95% CL). The upper limit on tensor modes from polarization data alone is a factor of two lower with the seven-year data than it was using the five-year data. The data remain consistent with the simple {Lambda}CDM model: the best-fit TT spectrum has an effective {chi}{sup 2} of 1227 for 1170 degrees of freedom, with a probability to exceed of 9.6%. The allowable volume in the six-dimensional space of {Lambda}CDM parameters has been reduced by a factor of 1.5 relative to the five-year volume, while the {Lambda}CDM model that allows for tensor modes and a running scalar spectral index has a factor of three lower volume when fit to the seven-year data. We test the parameter recovery process for bias and find that the scalar spectral index, n{sub s} , is biased high, but only by 0.09{sigma}, while the remaining parameters are biased by <0.15{sigma}. The improvement in the third peak measurement leads to tighter lower limits from WMAP on the number of relativistic degrees of freedom (e.g., neutrinos) in the early universe: N{sub eff}>2.7(95%CL). Also, using WMAP data alone, the primordial helium mass fraction is found to be Y{sub He} = 0.28{sup +0.14}{sub -0.15}, and with data from higher-resolution cosmic microwave background experiments included, we now establish the existence of pre-stellar helium at >3{sigma}. These new WMAP measurements provide important tests of big bang cosmology.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2012

EQUATION OF STATE FOR MASSIVE NEUTRON STARS

Using the relativistic Hartree-Fock approximation, we investigate the properties of neutron-star matter in detail. In the present calculation, we consider not only the tensor coupling of vector mesons to octet baryons and the form factors at interaction vertices but also the change in the internal (quark) structure of baryons in dense matter. The relativistic Hartree-Fock calculations are performed in two ways: one with coupling constants determined by SU(6) (quark model) symmetry and the other with coupling constants based on SU(3) (flavor) symmetry. For the latter case, we use the latest Nijmegen (ESC08) model. Then, it is very remarkable that the particle composition of the core matter in SU(3) symmetry is completely different from that in SU(6) symmetry. In SU(6) symmetry, all octet baryons appear in the density region below {approx}1.2 fm{sup -3}, while in the ESC08 model only the {Xi}{sup -} hyperon is produced. Furthermore, the medium modification of the internal baryon structure hardens the equation of state for the core matter. Taking all these effects into account, we can obtain the maximum neutron-star mass which is consistent with the recently observed mass, 1.97 {+-} 0.04 M{sub Sun} (PSR J1614-2230). We therefore conclude that the extension from SU(6) symmetry to SU(3) symmetry in meson-baryon couplings and the internal baryon structure variation in matter considerably enhance the mass of neutron stars. Furthermore, the effects of the form factor at the vertex and the Fock contribution, including the tensor coupling due to vector mesons, are indispensable for describing the core matter.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2010

Setting the scale of the pp and pp total cross sections using AdS/QCD

This paper is an addendum to earlier work where we computed the Pomeron contribution to pp and pp scattering in AdS/QCD. Our model for pp scattering in the Regge regime depends on four parameters: the slope and intercept of the Pomeron trajectory {alpha}{sub c}{sup '}, {alpha}{sub c}(0), a mass scale M{sub d}, which determines a form factor entering into matrix elements of the energy-momentum tensor, and a coupling {lambda}{sub P} between the lightest spin-two glueball and the proton, which sets the overall scale of the total cross section. Here we perform a more detailed computation of {lambda}{sub P} in the Sakai-Sugimoto model by using the construction of nucleons as instantons of the dual 5D gauge theory and an effective 5D fermion description of these nucleons which has been successfully used to compute a variety of nucleon-meson couplings. We find {lambda}{sub P,SS{approx_equal}}6.38 GeV{sup -1}, which is in reasonable agreement with the value {lambda}{sub P,fit}=8.28 GeV{sup -1} determined by fitting single Pomeron exchange to data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Exploration of Electrochemical Reactions at Organic–Inorganic Halide Perovskite Interfaces via Machine Learning in In Situ Time-of-Flight Secondary Ion Mass Spectrometry

The instability of hybrid organic–inorganic perovskite (HOIP) devices is one of the significant challenges preventing commercialization. Exploring these phenomena is severely limited by the complexity of the intrinsic electrochemistry of HOIPs, the presence of multiple volatile and mobile ionic species, and the possible role of environmentally induced reactions at surfaces and triple-phase junctions. Here, in situ studies of the electrochemistry of methylammonium lead bromide perovskite with the Au electrode interface are reported via light- and voltage-dependent time-of-flight secondary ion mass spectrometry (ToF-SIMS) imaging of lateral perovskite heterostructures. While ToF-SIMS allows for the visualization of the chemical composition along the surface and its evolution with light and electrical bias, the interpretation of the multidimensional data obtained is often limited due to strong correlations between chemical signatures and the need to track multiple peaks at once. Here, a machine learning workflow combining the Hough transform and non-negative matrix factorization and no-negative tensor decomposition is developed to avoid this limitation and extract salient features of associated chemical changes and to separate the light- and voltage-dependent dynamics. Finally, combining these in situ characterizations and the machine learning workflow provides comprehensive information on the chemical nature of moving species, ion accumulation, and interfacial electrochemical reactions in HOIP devices.

36 MATERIALS SCIENCE↗
NASA NTRS2003

Macroscopic Models of Radiative Transfer as Applied to Computation of the Radiation Field in the Solar Atmosphere

In this paper a new three dimensional half-moment model for radiative transfer is presented for a gray medium. It describes the evolution of the zeroth and first directional half moments of the radiative intensity. The closure is provided, similarly to Dubroca and Klar, by the maximum entropy concept. This work generalizes that model to three dimensions. The model presented here (the derivation being done in Ripoll and Wray, called the M(sup 1/2)(sub 1) model, is a hyperbolic system consisting of a total of eight equations in three dimensions, four equations for each direction. Each half model has the classical form of a macroscopic moment model in which the pressure tensor is constructed from the well-known Eddington tensor with a particular Eddington factor. Moreover, different source and border terms occur. The latter introduce couplings between the macroscopic and microscopic quantities and between the + and - streams, through the intensity in the plane perpendicular to the flux. The main theoretical application of the half moment model, treated in this paper, is its reduction to a full moment model, called M(sup +)(sub 1), for the particular but important case of a hot, opaque source radiating in a cold transparent (or semi-transparent) medium for very specific applications, such as stellar interiors or atmospheres, or combustion problems. The structure of the paper is as follows. In section 2, the model M(sup 1/2)(sub 1) is presented. In section 3, for the particular case of a hot, opaque source radiating into a cold medium, the half moment model is reduced to the M(sup +)(sub 1) model. In section 4, we first solve a simple and academic problem to validate the models, followed by a simplified solar atmosphere.

Ripoll, J.-F.↗
DOE OSTI2011

Charmless hadronic B decays into a tensor meson

Two-body charmless hadronic B decays involving a tensor meson in the final state are studied within the framework of QCD factorization (QCDF). Because of the G-parity of the tensor meson, both the chiral-even and chiral-odd two-parton light-cone distribution amplitudes of the tensor meson are antisymmetric under the interchange of momentum fractions of the quark and antiquark in the SU(3) limit. Our main results are: (i) In the naieve factorization approach, the decays such as B{sup -}{yields}K{sub 2}*{sup 0}{pi}{sup -} and B{sup 0}{yields}K{sub 2}*{sup -}{pi}{sup +} with a tensor meson emitted are prohibited because a tensor meson cannot be created from the local V-A or tensor current. Nevertheless, the decays receive nonfactorizable contributions in QCDF from vertex, penguin and hard spectator corrections. The experimental observation of B{sup -}{yields}K{sub 2}*{sup 0}{pi}{sup -} indicates the importance of nonfactorizable effects. (ii) For penguin-dominated B{yields}TP and TV decays, the predicted rates in naieve factorization are usually too small by 1 to 2 orders of magnitude. In QCDF, they are enhanced by power corrections from penguin annihilation and nonfactorizable contributions. (iii) The dominant penguin contributions to B{yields}K{sub 2}*{eta}{sup (')} arise from the processes: (a) b{yields}sss{yields}s{eta}{sub s} and (b) b{yields}sqq{yields}qK{sub 2}* with {eta}{sub q}=(uu+dd)/{radical}(2) and {eta}{sub s}=ss. The interference, constructive for K{sub 2}*{eta}{sup '} and destructive for K{sub 2}*{eta}, explains why {Gamma}(B{yields}K{sub 2}*{eta}{sup '})>>{Gamma}(B{yields}K{sub 2}*{eta}). (iv) We use the measured rates of B{yields}K{sub 2}*({omega},{phi}) to extract the penguin-annihilation parameters {rho}{sub A}{sup TV} and {rho}{sub A}{sup VT} and the observed longitudinal polarization fractions f{sub L}(K{sub 2}*{omega}) and f{sub L}(K{sub 2}*{phi}) to fix the phases {phi}{sub A}{sup VT} and {phi}{sub A}{sup TV}. (v) The experimental observation that f{sub T}/f{sub L}<<1 for B{yields}K{sub 2}*(1430){phi}, whereas f{sub T}/f{sub L}{approx}1 for B{yields}K{sub 2}*(1430){omega} with f{sub T} being the transverse polarization fraction, can be accommodated in QCDF, but it cannot be dynamically explained at first place. For penguin-dominated B{yields}TV decays, we find f{sub L}(K{sub 2}*{rho}){approx}f{sub L}(K{sub 2}*{omega}){approx}0.65, whereas f{sub L}(K*f{sub 2}){approx}0.93. It will be of great interest to measure f{sub L} for these modes to test QCDF. Theoretically, transverse polarization is expected to be small in tree-dominated B{yields}TV decays except for the a{sub 2}{sup -}{rho}{sup 0}, a{sub 2}{sup -}{rho}{sup +}, K{sub 2}*{sup 0}K*{sup -} and K{sub 2}*{sup 0}K*{sup 0} modes. (vi) For tree-dominated decays, their rates are usually very small except for the a{sub 2}{sup 0}({pi}{sup -},{rho}{sup -}), a{sub 2}{sup +}({pi}{sup -},{rho}{sup -}) and f{sub 2}({pi}{sup -},{rho}{sup -}) modes with branching fractions of order 10{sup -6} or even larger.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

spdlayers

Symmetric Positive Definite (SPD) enforcement layers for PyTorch. Regardless of the input, the output of these layers will always be a SPD tensor! The `Cholesky` layer uses a cholesky factorization to enforce SPD, and the `Eigen` layer uses an eigendecomposition to enforce SPD. Both layers take in some tensor of shape `[batch_size, input_shape]` and output a SPD tensor of shape`[batch_size, output_shape, output_shape]`. The relationship between input and output is defined by the following. ```python input_shape = sum([i for i in range(output_shape + 1)]) ``` The layers have no learnable parameters, and merely serve to transform a vector space to a SPD matrix space.

Jekel, CharlesF.↗
DOE OSTI2018

Parallel Nonnegative CP Decomposition of Dense Tensors

The CP tensor decomposition is a low-rank approximation of a tensor. We present a distributed-memory parallel algorithm and implementation of an alternating optimization method for computing a CP decomposition of dense tensors that can enforce nonnegativity of the computed low-rank factors. The principal task is to parallelize the Matricized-Tensor Times Khatri-Rao Product (MTTKRP) bottleneck subcomputation. The algorithm is computation efficient, using dimension trees to avoid redundant computation across MTTKRPs within the alternating method. Our approach is also communication efficient, using a data distribution and parallel algorithm across a multidimensional processor grid that can be tuned to minimize communication. We benchmark our software on synthetic as well as hyperspectral image and neuroscience dynamic functional connectivity data, demonstrating that our algorithm scales well to 100s of nodes (up to 4096 cores) and is faster and more general than the currently available parallel software.

Ballard, Grey↗
DOE OSTI2021

Multi-Area Model-Free State Estimation via Distributed Tensor Decomposition

This paper proposes a model-free method for distribution system state estimation based on tensor completion using canonical polyadic decomposition. In particular, we consider a setting where the network is divided into multiple areas. The measured physical quantities at buses located in the same area are processed by an area controller. A third-order tensor is constructed to collect these measured quantities. The measurements are analyzed locally to recover the full state information of the network. A closed-form iterative algorithm based on the alternating direction method of multipliers is developed to obtain the low-rank factors of the whole network state tensor where information exchange happens only between neighboring areas. To demonstrate the efficacy of the developed algorithm, numerical simulations are carried out using an IEEE test system.

alternating direction method of multipliers↗
NASA NTRS2009

Stress Concentration and Fracture at Inter-variant Boundaries in an Al-Li Alloy

Delamination fracture has limited the use of lightweight Al-Li alloys. Studies of secondary, delamination cracks in alloy 2090, L-T fracture toughness samples showed grain boundary failure between variants of the brass texture component. Although the adjacent texture variants, designated B(sub s1) and B(sub s2), behave similarly during rolling, their plastic responses to mechanical tests can be quite different. EBSD data from through-thickness scans were used to generate Taylor factor maps. When a combined boundary normal and shear tensor was used in the calculation, the delaminating grains showed the greatest Taylor Factor differences of any grain pairs. Kernel Average Misorientation (KAM) maps also showed damage accumulation on one side of the interface. Both of these are consistent with poor slip accommodation from a crystallographically softer grain to a harder one. Transmission electron microscopy was used to confirm the EBSD observations and to show the role of slip bands in the development of large, interfacial stress concentrations. A viewgraph presentation accompanies the provided abstract.

Crooks, Roy↗
DOE OSTI2025

A Linear-Complexity Tensor Butterfly Algorithm for Compressing High-Dimensional Oscillatory Integral Operators

This paper presents a multilevel tensor compression algorithm called tensor butterfly algorithm for efficiently representing large-scale and high-dimensional oscillatory integral operators, including Green's functions for wave equations and integral transforms such as Radon transforms and Fourier transforms. The proposed algorithm leverages a tensor extension of the so-called complementary low-rank property of existing matrix butterfly algorithms. The algorithm partitions the discretized integral operator tensor into subtensors of multiple levels and factorizes each subtensor at the middle level as a Tucker-type interpolative decomposition, whose factor matrices are formed in a multilevel fashion. For a d-dimensional (d > 1) integral operator discretized into a 2d-mode tensor with n2d entries, the overall CPU time and memory requirement scale as O(nd), in stark contrast to the O(nd log n) complexity of existing matrix algorithms such as matrix butterfly algorithms and fast Fourier transforms (FFTs), where n is the number of points per direction. When comparing with other tensor algorithms such as quantized tensor train (QTT), the proposed algorithm also shows superior CPU and memory performance for tensor contraction. Remarkably, the tensor butterfly algorithm can efficiently model high-frequency Green's function interactions between two unit cubes, each spanning 512 wavelengths per direction, which represents problems of scale over 512× larger than that existing butterfly algorithms can handle, with the same amount of computation resources. On the other hand, for a problem representing 64 wavelengths per direction, which is the largest size existing algebraic matrix algorithms can handle, our tensor butterfly algorithm exhibits 200x speedups and 30× memory reduction compared with existing ones. Moreover, the tensor butterfly algorithm also permits O(nd)-complexity FFTs and Radon transforms up to d = 6 dimensions.

Kielstra, P Michael↗
DOE OSTI2011

Visser's massive graviton bimetric theory revisited

A massive gravity theory was proposed by Visser in the late 1990s. This theory, based on a background metric b{sub {alpha}{beta}} and on an usual dynamical metric g{sub {alpha}{beta}} has the advantage of being free of ghosts as well as discontinuities present in other massive theories proposed in the past. In the present investigation, the equations of Visser's theory are revisited with particular care on the related conservation laws. It will be shown that a multiplicative factor is missing in the graviton tensor originally derived by Visser, which has no incidence on the weak field approach but becomes important in the strong field regime when, for instance, cosmological applications are considered. In this case, contrary to some previous claims found in the literature, we conclude that a nonstatic background metric is required in order to obtain a solution able to mimic the {Lambda}CDM cosmology.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

An experimental program with high duty-cycle polarized and unpolarized positron beams at Jefferson Lab

Positron beams, both polarized and unpolarized, are identified as important ingredients for the experimental programs at the next generation of lepton accelerators. In the context of the hadronic physics program at Jefferson Lab (JLab), positron beams are complementary, even essential, tools for a precise understanding of the electromagnetic structure of nucleons and nuclei, in both the elastic and deep-inelastic regimes. For instance, elastic scattering of polarized and unpolarized electrons and positrons from the nucleon enables a model independent determination of its electromagnetic form factors. Also, the deeply-virtual scattering of polarized and unpolarized electrons and positrons allows unambiguous separation of the different contributions to the cross section of the lepto-production of photons and of lepton-pairs, enabling an accurate determination of the nucleons and nuclei generalized parton distributions, and providing an access to the gravitational form factors of the energy-momentum tensor. Furthermore, positron beams offer the possibility of alternative tests of the Standard Model of particle physics through the search of a dark photon, the precise measurement of electroweak couplings, and the investigation of charged lepton flavor violation. This document discusses the perspectives of an experimental program with high duty-cycle positron beams at JLab.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2013

Physics-based seismic hazard analysis on petascale heterogeneous supercomputers, In: SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

We have developed a highly scalable and efficient GPU-based finite-difference code (AWP) for earthquake simulation that implements high throughput, memory locality, communication reduction and communication/computation overlap and achieves linear scalability on Cray XK7 Titan at ORNL and NCSA's Blue Waters system. We simulate realistic 0-10 Hz earthquake ground motions relevant to building engineering design using high-performance AWP. Moreover, we show that AWP provides a speedup by a factor of 110 in key strain tensor calculations critical to probabilistic seismic hazard analysis (PSHA). These performance improvements to critical scientific application software, coupled with improved co-scheduling capabilities of our workflow-managed systems, make a statewide hazard model a goal reachable with existing supercomputers. The performance improvements of GPU-based AWP are expected to save millions of core-hours over the next few years as physics-based seismic hazard analysis is developed using heterogeneous petascale supercomputers.

Computer Science↗
DOE OSTI2013

Chiral Dynamics and Transverse Peripheral Densities

We study the transverse charge and current densities in the periphery of the nucleon where they are governed by universal chiral dynamics. The large?distance asymptotics of the densities are derived from the spectral functions of the Dirac and Pauli form factors near the threshold at t = 4M?2 obtained from invariant chiral perturbation theory. In the equivalent light?front formulation, the peripheral densities arise from large?size ?N configurations in the nucleon?s light?front wave function obtained from the chiral Lagrangian. We compute the chiral component of the charge and current densities and explain their order?of?magnitude in the chiral expansion. We also discuss the role of ?? intermediate states in the large Nc limit and the heavy?baryon expansion of peripheral densities, and comment on progress extending an analogous analysis concerning transverse densities from form factors of the energy-momentum tensor and its connections with studies on GPDs.

Granados, Carlos↗
DOE OSTI2025

A two-loop four-point form factor at function level

Recently, the maximally-helicity-violating four-point form factor for the chiral stress-energy tensor in planar $\mathcal{N}$ = 4 super Yang-Mills was computed to three loops at the level of the symbol associated with multiple polylogarithms. It exhibits antipodal self-duality, or invariance under the combined action of a kinematic map and reversing the ordering of letters in the symbol. Here we lift the two-loop form factor from symbol level to function level. We provide an iterated representation of the function’s derivatives (coproducts). In order to do so, we find a three-parameter limit of the five-parameter phase space where the symbol’s letters are all rational. We also use function-level information about dihedral symmetries and the soft, collinear, and factorization limits, as well as limits governed by the form-factor operator product expansion (FFOPE). We provide plots of the remainder function on several kinematic slices, and show that the result is compatible with the FFOPE data. We further verify that antipodal self-duality is valid at two loops beyond the level of the symbol.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2011

ON DETERMINING THE SHAPE OF MATTER DISTRIBUTIONS

A basic property of objects, such as galaxies and halos that form in cosmological structure formation simulations, is their shape. Here, we critically investigate shape determination methods that are commonly used in the literature. It is found that using an enclosed integration volume and weight factors r{sup -2} and r{sup -2}{sub ell} (elliptical radius) for the contribution of each particle or volume element in the shape tensor leads to biased axis ratios and smoothing of details when calculating the local shape as a function of distance from the center. To determine the local shape of matter distributions as a function of distance for well-resolved objects (typically more than O(10{sup 4}) particles), we advocate a method that (1) uses an ellipsoidal shell (homoeoid) as an integration volume without any weight factors in the shape tensor and (2) removes subhalos.

79 ASTRONOMY AND ASTROPHYSICS↗