Search NASA⌕ Search

SEARCH · Search NASA

Results for “Decomposition algorithm”

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 793 records · Page 44

A flexible class of priors for orthonormal matrices with basis function-specific structure

Statistical modeling of high-dimensional matrix-valued data motivates the use of a low-rank representation that simultaneously summarizes key characteristics of the data and enables dimension reduction. Low-rank representations commonly factor the original data into the product of orthonormal basis functions and weights, where each basis function represents an independent feature of the data. However, the basis functions in these factorizations are typically computed using algorithmic methods that cannot quantify uncertainty or account for basis function correlation structure a priori. While there exist Bayesian methods that allow for a common correlation structure across basis functions, empirical examples motivate the need for basis function-specific dependence structure. We propose a prior distribution for orthonormal matrices that can explicitly model basis function-specific structure. The prior is used within a general probabilistic model for singular value decomposition to conduct posterior inference on the basis functions while accounting for measurement error and fixed effects. We discuss how the prior specification can be used for various scenarios and demonstrate favorable model properties through synthetic data examples. Finally, we apply our method to two-meter air temperature data from the Pacific Northwest, enhancing our understanding of the Earth system’s internal variability.

97 MATHEMATICS AND COMPUTING↗

Molecular dipole moment learning via rotationally equivariant derivative kernels in molecular-orbital-based machine learning

This study extends the accurate and transferable molecular-orbital-based machine learning (MOB-ML) approach to modeling the contribution of electron correlation to dipole moments at the cost of Hartree–Fock computations. A MOB pairwise decomposition of the correlation part of the dipole moment is applied, and these pair dipole moments could be further regressed as a universal function of MOs. The dipole MOB features consist of the energy MOB features and their responses to electric fields. An interpretable and rotationally equivariant derivative kernel for Gaussian process regression (GPR) is introduced to learn the dipole moment more efficiently. The proposed problem setup, feature design, and ML algorithm are shown to provide highly accurate models for both dipole moments and energies on water and 14 small molecules. To demonstrate the ability of MOB-ML to function as generalized density-matrix functionals for molecular dipole moments and energies of organic molecules, we further apply the proposed MOB-ML approach to train and test the molecules from the QM9 dataset. The application of local scalable GPR with Gaussian mixture model unsupervised clustering GPR scales up MOB-ML to a large-data regime while retaining the prediction accuracy. In addition, compared with the literature results, MOB-ML provides the best test mean absolute errors of 4.21 mD and 0.045 kcal/mol for dipole moment and energy models, respectively, when training on 110 000 QM9 molecules. The excellent transferability of the resulting QM9 models is also illustrated by the accurate predictions for four different series of peptides.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Pre-conditioned BFGS-based uncertainty quantification in elastic full-waveform inversion

SUMMARY Full-waveform inversion has become an essential technique for mapping geophysical subsurface structures. However, proper uncertainty quantification is often lacking in current applications. In theory, uncertainty quantification is related to the inverse Hessian (or the posterior covariance matrix). Even for common geophysical inverse problems its calculation is beyond the computational and storage capacities of the largest high-performance computing systems. In this study, we amend the Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm to perform uncertainty quantification for large-scale applications. For seismic inverse problems, the limited-memory BFGS (L-BFGS) method prevails as the most efficient quasi-Newton method. We aim to augment it further to obtain an approximate inverse Hessian for uncertainty quantification in FWI. To facilitate retrieval of the inverse Hessian, we combine BFGS (essentially a full-history L-BFGS) with randomized singular value decomposition to determine a low-rank approximation of the inverse Hessian. Setting the rank number equal to the number of iterations makes this solution efficient and memory-affordable even for large-scale problems. Furthermore, based on the Gauss–Newton method, we formulate different initial, diagonal Hessian matrices as pre-conditioners for the inverse scheme and compare their performances in elastic FWI applications. We highlight our approach with the elastic Marmousi benchmark model, demonstrating the applicability of pre-conditioned BFGS for large-scale FWI and uncertainty quantification.

58 GEOSCIENCES↗

ShiftNMFk 1.2

The ShiftNMFk1.2 code, or as we call it, GreenNMFk, represents a hybrid algorithm combining unsupervised adaptive machine learning and Green's function inverse method. GreenNMFk allows an efficient and high performance de-mixing and feature extraction of a multitude of nonnegative signals that change their shape propagating through the medium. The signals are mixed and recorded by a network of uncorrelated sensors. The code couples Non-negative Matrix Factorization (NMF) and inverse-analysis Green's functions method. GreenNMF synergistically performs decomposition of the recorded mixtures, finds the number of the unknown sources and uses the Green's function of the governing partial differential equation to identify the unknown sources and their charecteristics. GreenNMF can be applied directly to any problem controlled by a known partial-differential parabolic equation where mixtures of an unknown number of sources are measured at multiple locations. Full GreenNMFk method is a subject LANL U.S. Patent application S133364.000 August, 2017. The ShiftNMFk 1.2 version here is a toy version of this method that can work with a limited number of unknown sources (4 or less).

Alexandrov, BoianS.↗

Physics Informed Neural Networks as Computational Physics Emulators

This report is a brief overview and evaluation of Physics Informed Neural Networks (PINNs). Karniadakis and co-workers, e.g., Karniadakis et al. (2021) assert that the PINNs approach integrates seamlessly both data and mathematical physics models, even in partially understood, uncertain and high-dimensional contexts. They further claim that PINNs are effective and efficient for ill-posed and inverse problems, and when combined with domain decomposition, are scalable to large problems and a tool to discover hidden physics. While they demonstrate the capabilities in specific academic instances, their overarching claims about PINNs seem to be an overstatement, at least at the current time. We have briefly considered a few of the limitations of PINNs in this investigation. It is not clear to us if the PINNs approach can ever be competitive with approaches that use specialized algorithms to achieve high-accuracy solutions of governing equations and other techniques that can combine observational data with such solutions. As an example of the latter, consistent with the principles of Bayesian inference, data assimilation, or more generally data-model fusion, is a process that fuses observational data typically with a computational model that respects certain constraints such as conservation laws. For example, improvements in observational network combined with data assimilation have been key in improving weather predictions over the past four decades Kalnay (2003).

97 MATHEMATICS AND COMPUTING↗

Progress on Associate-Particle Imaging Algorithms, 2020

The present work describes progress on the development of imaging algorithms that use fast neutron signatures acquired using the associated-particle imaging (API) method. The present work complements ongoing work to develop neutron source and detector hardware to enable field inspection by investigating algorithms that are capable of discriminating between critical materials or extracting three-dimensional geometrical information from single-sided or transmission measurements. The present work is divided into three approaches:(1)Iterative reconstruction of inelastic gamma-ray emissions to perform three-dimensional time-of-flight imaging in a single view in either transmission or backscatter configurations. Iterative reconstruction enables image resolution better than the inherent TOF resolution.(2)Decomposition of registered neutron and x ray radiographs into an assumed material list for each pixel in the image.(3)Material identification using full spectral analysis that includes the emergent neutron and gamma ray energies, times, and angles.Progress for each approach is summarized for fiscal year 2020.

97 MATHEMATICS AND COMPUTING↗

GSoFa: Scalable Sparse Symbolic LU Factorization on GPUs

Decomposing a matrix $\mathbf {A}$ into a lower matrix $\mathbf {L}$ and an upper matrix $\mathbf {U}$, which is also known as LU decomposition, is an essential operation in numerical linear algebra. For a sparse matrix, LU decomposition often introduces more nonzero entries in the $\mathbf {L}$ and $\mathbf {U}$ factors than in the original matrix. A symbolic factorization step is needed to identify the nonzero structures of $\mathbf {L}$ and $\mathbf {U}$ matrices. Attracted by the enormous potentials of the Graphics Processing Units (GPUs), an array of efforts have surged to deploy various LU factorization steps except for the symbolic factorization, to the best of our knowledge, on GPUs. This article introduces gSoFa, the first GPU-based symbolic factorization design with the following three optimizations to enable scalable LU symbolic factorization for nonsymmetric pattern sparse matrices on GPUs. First, here we introduce a novel fine-grained parallel symbolic factorization algorithm that is well suited for the Single Instruction Multiple Thread (SIMT) architecture of GPUs. Second, we tailor supernode detection into a SIMT friendly process and strive to balance the workload, minimize the communication and saturate the GPU computing resources during supernode detection. Third, we introduce a three-pronged optimization to reduce the excessive space consumption problem faced by multi-source concurrent symbolic factorization. Taken together, gSoFa achieves up to 31× speedup from 1 to 44 Summit nodes (6 to 264 GPUs) and outperforms the state-of-the-art CPU project, on average, by 5×. Notably, gSoFa also achieves up to 47 percent of the peak memory throughput of a V100 GPU in the Summit Supercomputer.

97 MATHEMATICS AND COMPUTING↗

CORRECTING FOR INTERSTELLAR SCATTERING DELAY IN HIGH-PRECISION PULSAR TIMING: SIMULATION RESULTS

Light travel time changes due to gravitational waves (GWs) may be detected within the next decade through precision timing of millisecond pulsars. Removal of frequency-dependent interstellar medium (ISM) delays due to dispersion and scattering is a key issue in the detection process. Current timing algorithms routinely correct pulse times of arrival (TOAs) for time-variable delays due to cold plasma dispersion. However, none of the major pulsar timing groups correct for delays due to scattering from multi-path propagation in the ISM. Scattering introduces a frequency-dependent phase change in the signal that results in pulse broadening and arrival time delays. Any method to correct the TOA for interstellar propagation effects must be based on multi-frequency measurements that can effectively separate dispersion and scattering delay terms from frequency-independent perturbations such as those due to a GW. Cyclic spectroscopy, first described in an astronomical context by Demorest (2011), is a potentially powerful tool to assist in this multi-frequency decomposition. As a step toward a more comprehensive ISM propagation delay correction, we demonstrate through a simulation that we can accurately recover impulse response functions (IRFs), such as those that would be introduced by multi-path scattering, with a realistic signal-to-noise ratio (S/N). We demonstrate that timing precision is improved when scatter-corrected TOAs are used, under the assumptions of a high S/N and highly scattered signal. We also show that the effect of pulse-to-pulse “jitter” is not a serious problem for IRF reconstruction, at least for jitter levels comparable to those observed in several bright pulsars.

79 ASTRONOMY AND ASTROPHYSICS↗

Accurate simulation of direct laser acceleration in a laser wakefield accelerator

In a laser wakefield accelerator (LWFA), an intense laser pulse excites a plasma wave that traps and accelerates electrons to relativistic energies. When the pulse overlaps the accelerated electrons, it can enhance the energy gain through direct laser acceleration (DLA) by resonantly driving the betatron oscillations of the electrons in the plasma wave. The traditional particle-in-cell (PIC) algorithm, although often the tool of choice to study DLA, contains inherent errors due to numerical dispersion and the time staggering of the electric and magnetic fields. Furthermore, conventional PIC implementations cannot reliably disentangle the fields of the plasma wave and laser pulse, which obscures interpretation of the dominant acceleration mechanism. Here, a customized field solver that reduces errors from both numerical dispersion and time staggering is used in conjunction with a field decomposition into azimuthal modes to perform PIC simulations of DLA in an LWFA. Comparisons with traditional PIC methods, model equations, and experimental data show improved accuracy with the customized solver and convergence with an order-of-magnitude fewer cells. Furthermore, the azimuthal-mode decomposition reveals that the most energetic electrons receive comparable energy from DLA and LWFA.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

P3DFFT: A Framework for Parallel Computations of Fourier Transforms in Three Dimensions

Fourier and related transforms are a family of algorithms widely employed in diverse areas of computational science, notoriously difficult to scale on high-performance parallel computers with a large number of processing elements (cores). This paper introduces a popular software package called P3DFFT which implements fast Fourier transforms (FFTs) in three dimensions in a highly efficient and scalable way. It overcomes a well-known scalability bottleneck of three-dimensional (3D) FFT implementations by using two-dimensional domain decomposition. Designed for portable performance, P3DFFT achieves excellent timings for a number of systems and problem sizes. On a Cray XT5 system P3DFFT attains 45% efficiency in weak scaling from 128 to 65,536 computational cores. Library features include Fourier and Chebyshev transforms, Fortran and C interfaces, in- and out-of-place transforms, uneven data grids, and single and double precision. P3DFFT is available as open source at http://code.google.com/p/p3dfft/. This paper discusses P3DFFT implementation and performance in a way that helps guide the user in making optimal choices for parameters of their runs. Read More: https://epubs.siam.org/doi/10.1137/11082748X

Mathematics↗

Scalable time series change detection for biomass monitoring using gaussian process

Biomass monitoring, specifically detecting changes in the biomass or vegetation of a geographical region, is vital for studying the carbon cycle of the system and has significant implications in the context of understanding climate change and its impacts. Recently, several time series change detection methods have been proposed to identify land cover changes in temporal profiles (time series) of vegetation collected using remote sensing instruments. In this paper, we adapt Gaussian process regression to detect changes in such time series in an online fashion. While Gaussian process (GP) has been widely used as a kernel based learning method for regression and classification, their applicability to massive spatio-temporal data sets, such as remote sensing data, has been limited owing to the high computational costs involved. In this paper we address the scalability aspect of GP based time series change detection. Specifically, we exploit the special structure of the covariance matrix generated for GP analysis to come up with methods that can efficiently estimate the hyper-parameters associated with GP as well as identify changes in the time series while requiring a memory footprint which is linear in the size of input data, as compared to traditional method which involves solving a linear system of equations for the Choleksy decomposition of the quadratic sized covariance matrix. Experimental results show that our proposed method achieves significant speedups, as high as 1000, when processing long time series, while maintaining a small memory footprint. To further improve the computational complexity of the proposed method, we provide a parallel version which can concurrently process multiple input time series using the same set of hyper-parameters. The parallel version exploits the natural parallelization potential of the serial algorithm and is shown to perform significantly better than the serial version, with speedups as high as 10. Finally, we demonstrate the effectiveness of the proposed change detection method in identifying changes in Normalized Difference Vegetation Index (NDVI) data. Moreover, we show that the scalable solution is able to process NDVI data for the entire Iowa region significantly faster than the standard method.

Chandola, Varun↗

Implementing Molecular Dynamics for Hybrid High Performance Computers - 1. Short Range Forces

The use of accelerators such as general-purpose graphics processing units (GPGPUs) have become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power requirements. Hybrid high performance computers, machines with more than one type of floating-point processor, are now becoming more prevalent due to these advantages. In this work, we discuss several important issues in porting a large molecular dynamics code for use on parallel hybrid machines - 1) choosing a hybrid parallel decomposition that works on central processing units (CPUs) with distributed memory and accelerator cores with shared memory, 2) minimizing the amount of code that must be ported for efficient acceleration, 3) utilizing the available processing power from both many-core CPUs and accelerators, and 4) choosing a programming model for acceleration. We present our solution to each of these issues for short-range force calculation in the molecular dynamics package LAMMPS. We describe algorithms for efficient short range force calculation on hybrid high performance machines. We describe a new approach for dynamic load balancing of work between CPU and accelerator cores. We describe the Geryon library that allows a single code to compile with both CUDA and OpenCL for use on a variety of accelerators. Finally, we present results on a parallel test cluster containing 32 Fermi GPGPUs and 180 CPU cores.

43 PARTICLE ACCELERATORS↗

A dictionary learning algorithm for compression and reconstruction of streaming data in preset order

There has been an emerging interest in developing and applying dictionary learning (DL) to process massive datasets in the last decade. Many of these efforts, however, focus on employing DL to compress and extract a set of important features from data, while considering restoring the original data from this set a secondary goal. On the other hand, although several methods are able to process streaming data by updating the dictionary incrementally as new snapshots pass by, most of those algorithms are designed for the setting where the snapshots are randomly drawn from a probability distribution. In this paper, we present a new DL approach to compress and denoise massive dataset in real time, in which the data are streamed through in a preset order (instances are videos and temporal experimental data), so at any time, we can only observe a biased sample set of the whole data. Here, our approach incrementally builds up the dictionary in a relatively simple manner: if the new snapshot is adequately explained by the current dictionary, we perform a sparse coding to find its sparse representation; otherwise, we add the new snapshot to the dictionary, with a Gram-Schmidt process to maintain the orthogonality. To compress and denoise noisy datasets, we apply the denoising to the snapshot directly before sparse coding, which deviates from traditional dictionary learning approach that achieves denoising via sparse coding. Compared to full-batch matrix decomposition methods, where the whole data is kept in memory, and other mini-batch approaches, where unbiased sampling is often assumed, our approach has minimal requirement in data sampling and storage: i) each snapshot is only seen once then discarded, and ii) the snapshots are drawn in a preset order, so can be highly biased. Through experiments on climate simulations and scanning transmission electron microscopy (STEM) data, we demonstrate that the proposed approach performs competitively to those methods in data reconstruction and denoising.

97 MATHEMATICS AND COMPUTING↗

Unconstrained plastering : all-hexahedral mesh generation via advancing front geometry decomposition (2004-2008).

The generation of all-hexahedral finite element meshes has been an area of ongoing research for the past two decades and remains an open problem. Unconstrained plastering is a new method for generating all-hexahedral finite element meshes on arbitrary volumetric geometries. Starting from an unmeshed volume boundary, unconstrained plastering generates the interior mesh topology without the constraints of a pre-defined boundary mesh. Using advancing fronts, unconstrained plastering forms partially defined hexahedral dual sheets by decomposing the geometry into simple shapes, each of which can be meshed with simple meshing primitives. By breaking from the tradition of previous advancing-front algorithms, which start from pre-meshed boundary surfaces, unconstrained plastering demonstrates that for the tested geometries, high quality, boundary aligned, orientation insensitive, all-hexahedral meshes can be generated automatically without pre-meshing the boundary. Examples are given for meshes from both solid mechanics and geotechnical applications.

97 MATHEMATICS AND COMPUTING↗

Recognizing molecular patterns by machine learning: An agnostic structural definition of the hydrogen bond

The concept of chemical bonding can ultimately be seen as a rationalization of the recurring structural patterns observed in molecules and solids. Chemical intuition is nothing but the ability to recognize and predict such patterns, and how they transform into one another. Here, we discuss how to use a computer to identify atomic patterns automatically, so as to provide an algorithmic definition of a bond based solely on structural information. We concentrate in particular on hydrogen bonding – a central concept to our understanding of the physical chemistry of water, biological systems, and many technologically important materials. Since the hydrogen bond is a somewhat fuzzy entity that covers a broad range of energies and distances, many different criteria have been proposed and used over the years, based either on sophisticate electronic structure calculations followed by an energy decomposition analysis, or on somewhat arbitrary choices of a range of structural parameters that is deemed to correspond to a hydrogen-bonded configuration. We introduce here a definition that is univocal, unbiased, and adaptive, based on our machine-learning analysis of an atomistic simulation. The strategy we propose could be easily adapted to similar scenarios, where one has to recognize or classify structural patterns in a material or chemical compound.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

TH-C-18A-02: Machine Learning and STAPLE Based Simultaneous Longitudinal Segmentation of Bone and Marrow Structures From Dual Energy CT

Purpose: To develop a fully-automatic longitudinal bone and marrow segmentation method in the pelvic region from dual energy computed tomography (DECT). Methods: We developed a two-step automatic bone and marrow segmentation method for simultaneous longitudinal evaluation of patients with metastatic bone disease using dual energy CT (DECT). Our approach transforms the DECT images into a multi-material decomposition (MMD) model that represents the voxels as a mixture of multiple materials. A support vector machine (SVM) was trained using a single scan. In the first step of the longitudinal segmentation the trained SVM model detects bone and marrow structures on all available longitudinal scans. Segmentation is further refined through active contour segmentation. In the second step, the segmentations from the individual scans are merged by employing the simultaneous truth and performance level estimation (STAPLE) algorithm. The scans are registered using affine and deformable registration. We found that our approach improves the segmentation in all the scans under reliable registration performance between the same scans. Improving registration was not under the scope of this work. Results: We applied our approach to segment bone and marrow in DECT scans in the pelvic regions for multiple patients. Each patient had three to five follow up scans. All the patients in the analysis had artificial metal prostheses which introduced challenges for the registration. Our algorithm achieved reasonable accurate segmentation despite the presence of metal artifacts and high-density oral contrast in neighboring structures. Our approach obtained an overall segmentation accuracy of 80% using DICE metric. Conclusion: We developed a two-step automatic longitudinal segmentation technique for bone and marrow region structures in the pelvic areas from dual energy CT. Our approach achieves robust segmentation despite the presence of confounding structures with similar intensities as the bone and marrow regions due to uptake of oral contrast and metal artifacts.

60 APPLIED LIFE SCIENCES↗

ALESQP: An Augmented Lagrangian Equality-Constrained SQP Method for Optimization with General Constraints

Here we present a new algorithm for infinite-dimensional optimization with general constraints, called ALESQP. In short, ALESQP is an augmented Lagrangian method that penalizes inequality constraints and solves equality-constrained nonlinear optimization subproblems at every iteration. The subproblems are solved using a matrix-free trust-region sequential quadratic programming (SQP) method that takes advantage of iterative, i.e., inexact linear solvers, and is suitable for large-scale applications. A key feature of ALESQP is a constraint decomposition strategy that allows it to exploit problem-specific variable scalings and inner products. We analyze convergence of ALESQP under different assumptions. We show that strong accumulation points are stationary. Consequently, in finite dimensions ALESQP converges to a stationary point. In infinite dimensions we establish that weak accumulation points are feasible in many practical situations. Under additional assumptions we show that weak accumulation points are stationary. We present several infinite-dimensional examples where ALESQP shows remarkable discretization-independent performance in all of its iterative components, requiring a modest number of iterations to meet constraint tolerances at the level of machine precision. Also, we demonstrate a fully matrix-free solution of an infinite-dimensional problem with nonlinear inequality constraints.

97 MATHEMATICS AND COMPUTING↗

Synergies from using higher order symplectic decompositions both for ordinary differential equations and quantum Monte Carlo methods

Decomposition methods of the Suzuki-Trotter type of various orders have been derived in different fields. Applying them both to classical ordinary differential equations (ODEs) and quantum systems allows to judge their effectiveness and gives new insights for many body quantum mechanics where reference data are scarce. Further, based on data for 6 × 6 system we conclude that sampling with sign (minus-sign problem) is probably detrimental to the accuracy of fermionic simulations with determinant algorithms.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗