Search NASA⌕ Search

SEARCH · Search NASA

Results for “tensor methods”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 487 records · Page 27

Generalized Recursion Relations for Correlators in the Gauge-Gravity Correspondence

We show that a generalization of the Britto-Cachazo-Feng-Witten recursion relations gives a new and efficient method of computing correlation functions of the stress tensor or conserved currents in conformal field theories with an (d+1)-dimensional anti-de Sitter space dual, for d{>=}4, in the limit where the bulk theory is approximated by tree-level Yang-Mills theory or gravity. In supersymmetric theories, additional correlators of operators that live in the same multiplet as a conserved current or stress tensor can be computed by these means.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A High-Throughput Solver for Marginalized Graph Kernels on GPU

Here, we present the design and optimization of a solver for efficient and high-throughput computation of the marginalized graph kernel on General Purpose GPUs. The graph kernel is computed using the conjugate gradient method to solve a generalized Laplacian of the tensor product between a pair of graphs. To cope with the large gap between the instruction throughput and the memory bandwidth of the GPUs, our solver forms the graph tensor product on-the-fly without storing it in memory. This is achieved by using threads in a warp cooperatively to stream the adjacency and edge label matrices of individual graphs by small square matrix blocks called tiles, which are then staged in registers and the shared memory for later reuse. Warps across a thread block can further share tiles via the shared memory to increase data reuse. We exploit the sparsity of the graphs hierarchically by storing only non-empty tiles using a coordinate format and nonzero elements within each tile using bitmaps. We propose a new partition-based reordering algorithm for aggregating nonzero elements of the graphs into fewer but denser tiles to further exploit sparsity. We carry out extensive theoretical analyses on the graph tensor product primitives for tiles of various density and evaluate their performance on synthetic and real-world datasets. Our solver delivers three to four orders of magnitude speedup over existing CPU-based solvers such as GraKeL and GraphKernels. The capability of the solver enables kernel-based learning tasks at unprecedented scales.

97 MATHEMATICS AND COMPUTING↗

Preserving Superconvergence of Spectral Elements for Curved Domains [Slides]

Finite Element Methods (FEM) and Spectral Element Methods (SEM) are crucial for solving partial differential equations (PDEs) on complex geometries. SEM offers superior accuracy due to potential superconvergence for simple domains. Challenges persist for domains with curved boundaries, restricting SEM’s advantages in real-world applications. A proposed solution is the introduction of a novel strategy to enhance accuracy and maintain superconvergence of SEM in curved domains. The strategy includes a mesh-generation procedure with geometrically refined elements near curved boundaries and a post-processing phase using the Adaptive Extended Stencil Finite Element Method (AES-FEM). The method, named AES-FEM post-processed Spectral Element Method (ApSEM), aligns the accuracy of non-tensor-product elements with superconvergent spectral elements.

97 MATHEMATICS AND COMPUTING↗

General-Purpose Unsupervised Cyber Anomaly Detection via Non-Negative Tensor Factorization

Distinguishing malicious anomalous activities from unusual but benign activities is a fundamental challenge for cyber defenders. Prior studies have shown that statistical user behavior analysis yields accurate detections by learning behavior profiles from observed user activity. These unsupervised models are able to generalize to unseen types of attacks by detecting deviations from normal behavior, without knowledge of specific attack signatures. However, approaches proposed to date based on probabilistic matrix factorization are limited by the information conveyed in a two-dimensional space. Non-negative tensor factorization, on the other hand, is a powerful unsupervised machine learning method that naturally models multi-dimensional data, capturing complex and multi-faceted details of behavior profiles. Herein, our new unsupervised statistical anomaly detection methodology matches or surpasses state-of-the-art supervised learning baselines across several challenging and diverse cyber application areas, including detection of compromised user credentials, botnets, spam e-mails, and fraudulent credit card transactions.

97 MATHEMATICS AND COMPUTING↗

Magnification Bias in Gravitational Arc Statistics

The statistics of gravitational arcs in galaxy clusters is a powerful probe of cluster structure and may provide complementary cosmological constraints. Despite recent progresses, discrepancies still remain among modelling and observations of arc abundance, specially regarding the redshift distribution of strong lensing clusters. Besides, fast "semi-analytic" methods still have to incorporate the success obtained with simulations. In this paper we discuss the contribution of the magnification in gravitational arc statistics. Although lensing conserves surface brightness, the magnification increases the signal-to-noise ratio of the arcs, enhancing their detectability. We present an approach to include this and other observational effects in semi-analytic calculations for arc statistics. The cross section for arc formation ({\sigma}) is computed through a semi-analytic method based on the ratio of the eigenvalues of the magnification tensor. Using this approach we obtained the scaling of {\sigma} with respect to the magnification, and other parameters, allowing for a fast computation of the cross section. We apply this method to evaluate the expected number of arcs per cluster using an elliptical Navarro--Frenk--White matter distribution. Our results show that the magnification has a strong effect on the arc abundance, enhancing the fraction of arcs, moving the peak of the arc fraction to higher redshifts, and softening its decrease at high redshifts. We argue that the effect of magnification should be included in arc statistics modelling and that it could help to reconcile arcs statistics predictions with the observational data.

79 ASTRONOMY AND ASTROPHYSICS↗

Gravitational lens optical scalars in terms of energy-momentum distributions

This is a general work on gravitational lensing. We present new expressions for the optical scalars and the deflection angle in terms of the energy-momentum tensor components of matter distributions. Our work generalizes standard references in the literature where normally stringent assumptions are made on the sources. The new expressions are manifestly gauge invariant, since they are presented in terms of curvature components. We also present a method of approximation for solving the lens equations, that can be applied to any order.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Rapid sampling of stochastic displacements in Brownian dynamics simulations

Here, we present a new method for sampling stochastic displacements in Brownian Dynamics (BD) simulations of colloidal scale particles. The method relies on a new formulation for Ewald summation of the Rotne-Prager-Yamakawa (RPY) tensor, which guarantees that the real-space and wave-space contributions to the tensor are independently symmetric and positive-definite for all possible particle configurations. Brownian displacements are drawn from a superposition of two independent samples: a wave-space (far-field or long-ranged) contribution, computed using techniques from fluctuating hydrodynamics and non-uniform fast Fourier transforms; and a real-space (near-field or short-ranged) correction, computed using a Krylov subspace method. The combined computational complexity of drawing these two independent samples scales linearly with the number of particles. The proposed method circumvents the super-linear scaling exhibited by all known iterative sampling methods applied directly to the RPY tensor that results from the power law growth of the condition number of tensor with the number of particles. For geometrically dense microstructures (fractal dimension equal three), the performance is independent of volume fraction, while for tenuous microstructures (fractal dimension less than three), such as gels and polymer solutions, the performance improves with decreasing volume fraction. This is in stark contrast with other related linear-scaling methods such as the force coupling method and the fluctuating immersed boundary method, for which performance degrades with decreasing volume fraction. Calculations for hard sphere dispersions and colloidal gels are illustrated and used to explore the role of microstructure on performance of the algorithm. In practice, the logarithmic part of the predicted scaling is not observed and the algorithm scales linearly for up to 4×10 6 particles, obtaining speed ups of over an order of magnitude over existing iterative methods, and making the cost of computing Brownian displacements comparable to the cost of computing deterministic displacements in BD simulations. Finally, a high-performance implementation employing non-uniform fast Fourier transforms implemented on graphics processing units and integrated with the software package HOOMD-blue is used for benchmarking.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Discrete generative diffusion models without stochastic differential equations: A tensor network approach

Diffusion models (DMs) are a class of generative machine learning methods that sample a target distribution by transforming samples of a trivial (often Gaussian) distribution using a learned stochastic differential equation. In standard DMs, this is done by learning a “score function” that reverses the effect of adding diffusive noise to the distribution of interest. Here we consider the generalisation of DMs to lattice systems with discrete degrees of freedom, and where noise is added via Markov chain jump dynamics. We show how to use tensor networks (TNs) to efficiently define and sample such “discrete diffusion models” (DDMs) without explicitly having to solve a stochastic differential equation. We show the following: (i) by parametrising the data and evolution operators as TNs, the denoising dynamics can be represented exactly; (ii) the auto-regressive nature of TNs allows to generate samples efficiently and without bias; (iii) for sampling Boltzmann-like distributions, TNs allow to construct an efficient learning scheme that integrates well with Monte Carlo. We illustrate this approach to study the equilibrium of two models with non-trivial thermodynamics, the d = 1 constrained Fredkin chain and the d = 2 Ising model. Published by the American Physical Society 2025

Causer, Luke (ORCID:0000000194243473)↗

Tunable Tensor Voting Improves Grouping of Membrane-Bound Macromolecules

Membrane-bound macromolecules are responsible for structural support and mediation of cell-cell adhesion in tissues. Quantitative analysis of these macromolecules provides morphological indices for damage or loss of tissue, for example as a result of exogenous stimuli. From an optical point of view, a membrane signal may have nonuniform intensity around the cell boundary, be punctate or diffused, and may even be perceptual at certain locations along the boundary. In this paper, a method for the detection and grouping of punctate, diffuse curvilinear signals is proposed. Our work builds upon the tensor voting and the iterative voting frameworks to propose an efficient method to detect and refine perceptually interesting curvilinear structures in images. The novelty of our method lies on the idea of iteratively tuning the tensor voting fields, which allows the concentration of the votes only over areas of interest. We validate the utility of our system with synthetic and annotated real data. The effectiveness of the tunable tensor voting is demonstrated on complex phenotypic signals that are representative of membrane-bound macromolecular structures.

60 APPLIED LIFE SCIENCES↗

A numerical-homogenization based phase-field fracture modeling of linear elastic heterogeneous porous media

Most porous media, such as geomaterials and biomaterials are highly heterogeneous in nature, and they contain large variations of microscopic pore structures, such as pore sizes, pore distribution, and pore shapes. The oscillation of microscopic structures is a substantial challenge in theoretical characterization and is usually ignored in continuous modeling. However, mechanical behavior of porous media such as deformation and failure, are essentially impacted by the microscopic heterogeneity which needs to be considered in modeling a porous media. Here, this research proposes a numerical modeling framework with a capability to investigate the effect of microscopic heterogeneity on the macroscopic fracture behavior in porous media by using a numerical homogenization technique, combined with the phase-field fracture modeling method. This numerical modeling strategy computes a homogenized elasticity tensor based on microscopic heterogeneous pore structures heterogenous porous domain by solving boundary value problems at microscopic domain. The strain energy and subsequent propagation of macroscopic fractures will be updated using homogenized stiffness information. Using this numerical scheme, the microscopic pore structure’s impact on the fracture behavior through the homogenized elastic tensor will be taken into account. This multiscale technique is benchmarked against classical problems. Finally, the results highlight the importance of the underlying pore structure and reveal that both fracture strength and propagation path can be influenced by the microscopic heterogeneity.

36 MATERIALS SCIENCE↗

Quantum gravitational correction to the Hawking temperature from the Lemaitre-Tolman-Bondi model

We solve the quantum constraint equations of the Lemaitre-Tolman-Bondi model in a semiclassical approximation in which an expansion is performed with respect to the Planck length. We recover in this way the standard expression for the Hawking temperature as well as its first quantum gravitational correction. We then interpret this correction in terms of the one-loop trace anomaly of the energy-momentum tensor and thereby make contact with earlier work on quantum black holes.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Markov Jump Processes Approximating a Non-Symmetric Generalized Diffusion

Consider a non-symmetric generalized diffusion X( Dot-Operator ) in Double-Struck-Capital-R {sup d} determined by the differential operator A(x) = -{Sigma}{sub ij} {partial_derivative}{sub i}a{sub ij}(x){partial_derivative}{sub j} + {Sigma}{sub i} b{sub i}(x){partial_derivative}{sub i}. In this paper the diffusion process is approximated by Markov jump processes X{sub n}( Dot-Operator ), in homogeneous and isotropic grids G{sub n} Subset-Of Double-Struck-Capital-R {sup d}, which converge in distribution in the Skorokhod space D([0,{infinity}), Double-Struck-Capital-R {sup d}) to the diffusion X( Dot-Operator ). The generators of X{sub n}( Dot-Operator ) are constructed explicitly. Due to the homogeneity and isotropy of grids, the proposed method for d{>=}3 can be applied to processes for which the diffusion tensor {l_brace}a{sub ij}(x){r_brace}{sub 11}{sup dd} fulfills an additional condition. The proposed construction offers a simple method for simulation of sample paths of non-symmetric generalized diffusion. Simulations are carried out in terms of jump processes X{sub n}( Dot-Operator ). For piece-wise constant functions a{sub ij} on Double-Struck-Capital-R {sup d} and piece-wise continuous functions a{sub ij} on Double-Struck-Capital-R {sup 2} the construction and principal algorithm are described enabling an easy implementation into a computer code.

97 MATHEMATICS AND COMPUTING↗

Nonlinear Acceleration Methods for Even-Parity Neutron Transport

Convergence acceleration methods for even-parity transport were developed that have the potential to speed up transport calculations and provide a natural avenue for an implicitly coupled multiphysics code. An investigation was performed into the acceleration properties of the introduction of a nonlinear quasi-diffusion-like tensor in linear and nonlinear solution schemes. Using the tensor reduced matrix as a preconditioner for the conjugate gradients method proves highly efficient and effective. The results for the linear and nonlinear case serve as the basis for further research into the application in a full three-dimensional spherical-harmonics even-parity transport code. Once moved into the nonlinear solution scheme, the implicit coupling of the convergence accelerated transport method into codes for other physics can be done seamlessly, providing an efficient, fully implicitly coupled multiphysics code with high order transport.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Reduced scaling extended multi-state CASPT2 (XMS-CASPT2) using supporting subspaces and tensor hyper-contraction

We present a reduced scaling formulation of the extended multi-state CASPT2 (XMS-CASPT2) method, which is based on our recently developed state-specific CASPT2 (SS-CASPT2) formulation using supporting subspaces and tensor hyper-contraction. By using these two techniques, the off-diagonal elements of the effective Hamiltonian can be computed with only O(N 3 ) operations and O(N 2 ) memory, where N is the number of basis functions. Furthermore, this limits the overall computational scaling to O(N 4 ) operations and O(N 2 ) memory. Thus, excited states can now be obtained at the same reduced (relative to previous algorithms) scaling we achieved for SS-CASPT2.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

3D Representative Volume Element Reconstruction of Fiber Composites via Orientation Tensor and Substructure Features

To provide a seamless integration of manufacturing processing simulation and fiber microstructure modeling, two new stochastic 3D microstructure reconstruction methods are proposed for two types of random fiber composites: random short fiber composites, and Sheet Molding Compounds (SMC) chopped fiber composites. A Random Sequential Adsorption (RSA) algorithm is first developed to embed statistical orientation information into 3D RVE reconstruction of random short fiber composites. For the SMC composites, an optimized Voronoi diagram based approach is developed for capturing the substructure features of SMC chopped fiber composites. The proposed methods are distinguished from other reconstruction works by providing a way of integrating statistical information (fiber orientation tensor) obtained from material processing simulation, as well as capturing the multiscale substructures of the SMC composites.

Li, Yi↗

Calculations of Solid-State 43Ca NMR Parameters: A Comparison of Periodic and Cluster Approaches and an Evaluation of DFT Functionals

We present a computational study of magnetic-shielding and quadrupolar-coupling tensors of 43Ca sites in crystalline solids. A comparison between periodic and cluster-based approaches for modeling solid-state interactions demonstrates that clusterbased approaches are suitable for predicting 43Ca NMR parameters. Several model chemistries, including Hartree–Fock theory and 17 DFT approximations (SVWN, CA-PZ, PBE, PBE0, PW91, B3PW91, rPBE, PBEsol, WC, PKZB, BMK, M06-L, M06, M06-2X, M06-HF, TPSS, and TPSSh), are evaluated for the prediction of 43Ca NMR parameters. Convergence of NMR parameters with respect to basis sets of the form cc-pVXZ (X5D, T, Q) is also evaluated. All DFT methods lead to substantial, and frequently systematic, overestimations of experimental chemical shifts. Hartree–Fock calculations outperform all DFT methods for the prediction of 43Ca chemical-shift tensors. VC 2017 Wiley Periodicals, Inc. DOI: 10.1002/jcc.24763

Holmes, Sean T.↗

NWChem: Quantum Chemistry Simulations at Scale

Methods based on quantum mechanics equations have been developed since the 1930's with the purpose of accurately studying the electronic structure of molecules. However, it is only during the last two decades that intense development of new computational algorithms has opened the possibility of performing accurate simulations of challenging molecular processes with high-order many-body methods. A wealth of evidence indicates that the proper inclusion of instantaneous interactions between electrons (or the so-called electron correlation effects) is indispensable for the accurate characterization of chemical reactivity, molecular properties, and interactions of light with matter. The availability of reliable methods for benchmarking of medium-size molecular systems provides also a unique chance to propagate high-level accuracy across spatial scales through the multiscale methodologies. Some of these methods have potential to utilize computational resources in an effi*cient way since they are characterized by high numerical complexity and appropriate level of data granularity, which can be effi*ciently distributed over multi-processor architectures. The broad spectrum of coupled cluster (CC) methods falls into this class of methodologies. Several recent CC implementations clearly demonstrated the scalability of CC formalisms on architectures composed of hundreds thousand computational cores. In this context NWChem provides a collection of Tensor Contraction Engine (TCE) generated parallel implementations of various coupled cluster methods capable of taking advantage of many thousand of cores on leadership class parallel architectures.

chemistry↗

Quantum annealing algorithms for Boolean tensor networks

Abstract Quantum annealers manufactured by D-Wave Systems, Inc., are computational devices capable of finding high-quality heuristic solutions of NP-hard problems. In this contribution, we explore the potential and effectiveness of such quantum annealers for computing Boolean tensor networks. Tensors offer a natural way to model high-dimensional data commonplace in many scientific fields, and representing a binary tensor as a Boolean tensor network is the task of expressing a tensor containing categorical (i.e., $$\{0, 1\}$$ { 0 , 1 } ) values as a product of low dimensional binary tensors. A Boolean tensor network is computed by Boolean tensor decomposition, and it is usually not exact. The aim of such decomposition is to minimize the given distance measure between the high-dimensional input tensor and the product of lower-dimensional (usually three-dimensional) tensors and matrices representing the tensor network. In this paper, we introduce and analyze three general algorithms for Boolean tensor networks: Tucker, Tensor Train, and Hierarchical Tucker networks. The computation of a Boolean tensor network is reduced to a sequence of Boolean matrix factorizations, which we show can be expressed as a quadratic unconstrained binary optimization problem suitable for solving on a quantum annealer. By using a novel method we introduce called parallel quantum annealing, we demonstrate that Boolean tensor’s with up to millions of elements can be decomposed efficiently using a DWave 2000Q quantum annealer.

97 MATHEMATICS AND COMPUTING↗