Search NASA⌕ Search

SEARCH · Search NASA

Results for “hierarchical 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 379 records · Page 21

TRIPLE-STAR CANDIDATES AMONG THE KEPLER BINARIES

We present the results of a search through the photometric database of Kepler eclipsing binaries looking for evidence of hierarchical triple-star systems. The presence of a third star orbiting the binary can be inferred from eclipse timing variations. We apply a simple algorithm in an automated determination of the eclipse times for all 2157 binaries. The ''calculated'' eclipse times, based on a constant period model, are subtracted from those observed. The resulting O - C (observed minus calculated times) curves are then visually inspected for periodicities in order to find triple-star candidates. After eliminating false positives due to the beat frequency between the {approx}1/2 hr Kepler cadence and the binary period, 39 candidate triple systems were identified. The periodic O - C curves for these candidates were then fit for contributions from both the classical Roemer delay and so-called physical delay, in an attempt to extract a number of the system parameters of the triple. We discuss the limitations of the information that can be inferred from these O - C curves without further supplemental input, e.g., ground-based spectroscopy. Based on the limited range of orbital periods for the triple-star systems to which this search is sensitive, we can extrapolate to estimate that at least 20% of all close binaries have tertiary companions.

79 ASTRONOMY AND ASTROPHYSICS↗

An Efficient Multicore Implementation of a Novel HSS-Structured Multifrontal Solver Using Randomized Sampling

Here, we present a sparse linear system solver that is based on a multifrontal variant of Gaussian elimination and exploits low-rank approximation of the resulting dense frontal matrices. We use hierarchically semiseparable (HSS) matrices, which have low-rank off-diagonal blocks, to approximate the frontal matrices. For HSS matrix construction, a randomized sampling algorithm is used together with interpolative decompositions. The combination of the randomized compression with a fast ULV HSS factoriz ation leads to a solver with lower computational complexity than the standard multifrontal method for many applications, resulting in speedups up to 7 fold for problems in our test suite. The implementation targets many-core systems by using task parallelism with dynamic runtime scheduling. Numerical experiments show performance improvements over state-of-the-art sparse direct solvers. The implementation achieves high performance and good scalability on a range of modern shared memory parallel systems, including the Intel Xeon Phi (MIC). The code is part of a software package called STRUMPACK - STRUctured Matrices PACKage, which also has a distributed memory component for dense rank-structured matrices.

97 MATHEMATICS AND COMPUTING↗

Tool for Ranking Research Options

Tool for Research Enhancement Decision Support (TREDS) is a computer program developed to assist managers in ranking options for research aboard the International Space Station (ISS). It could likely also be adapted to perform similar decision-support functions in industrial and academic settings. TREDS provides a ranking of the options, based on a quantifiable assessment of all the relevant programmatic decision factors of benefit, cost, and risk. The computation of the benefit for each option is based on a figure of merit (FOM) for ISS research capacity that incorporates both quantitative and qualitative inputs. Qualitative inputs are gathered and partly quantified by use of the time-tested analytical hierarchical process and used to set weighting factors in the FOM corresponding to priorities determined by the cognizant decision maker(s). Then by use of algorithms developed specifically for this application, TREDS adjusts the projected benefit for each option on the basis of levels of technical implementation, cost, and schedule risk. Based partly on Excel spreadsheets, TREDS provides screens for entering cost, benefit, and risk information. Drop-down boxes are provided for entry of qualitative information. TREDS produces graphical output in multiple formats that can be tailored by users.

Ortiz, James N.↗

Alert Classification for the ALeRCE Broker System: The Light Curve Classifier

We present the first version of the Automatic Learning for the Rapid Classification of Events (ALeRCE) broker light curve classifier. ALeRCE is currently processing the Zwicky Transient Facility (ZTF) alert stream, in preparation for the Vera C. Rubin Observatory. The ALeRCE light curve classifier uses variability features computed from the ZTF alert stream and colors obtained from AllWISE and ZTF photometry. We apply a balanced random forest algorithm with a two-level scheme where the top level classifies each source as periodic, stochastic, or transient, and the bottom level further resolves each of these hierarchical classes among 15 total classes. This classifier corresponds to the first attempt to classify multiple classes of stochastic variables (including core- and host-dominated active galactic nuclei, blazars, young stellar objects, and cataclysmic variables) in addition to different classes of periodic and transient sources, using real data. We created a labeled set using various public catalogs (such as the Catalina Surveys and Gaia DR2 variable stars catalogs, and the Million Quasars catalog), and we classify all objects with ≥6 g-band or ≥6 r-band detections in ZTF (868,371 sources as of 2020 June 9), providing updated classifications for sources with new alerts every day. For the top level we obtain macro-averaged precision and recall scores of 0.96 and 0.99, respectively, and for the bottom level we obtain macro-averaged precision and recall scores of 0.57 and 0.76, respectively. Updated classifications from the light curve classifier can be found at the ALeRCE Explorer website (http://alerce.online).

47 OTHER INSTRUMENTATION↗

TYPE Ia SUPERNOVA LIGHT-CURVE INFERENCE: HIERARCHICAL BAYESIAN ANALYSIS IN THE NEAR-INFRARED

We present a comprehensive statistical analysis of the properties of Type Ia supernova (SN Ia) light curves in the near-infrared using recent data from Peters Automated InfraRed Imaging TELescope and the literature. We construct a hierarchical Bayesian framework, incorporating several uncertainties including photometric error, peculiar velocities, dust extinction, and intrinsic variations, for principled and coherent statistical inference. SN Ia light-curve inferences are drawn from the global posterior probability of parameters describing both individual supernovae and the population conditioned on the entire SN Ia NIR data set. The logical structure of the hierarchical model is represented by a directed acyclic graph. Fully Bayesian analysis of the model and data is enabled by an efficient Markov Chain Monte Carlo algorithm exploiting the conditional probabilistic structure using Gibbs sampling. We apply this framework to the JHK{sub s} SN Ia light-curve data. A new light-curve model captures the observed J-band light-curve shape variations. The marginal intrinsic variances in peak absolute magnitudes are sigma(M{sub J}) = 0.17 +- 0.03, sigma(M{sub H}) = 0.11 +- 0.03, and sigma(M{sub Ks}) = 0.19 +- 0.04. We describe the first quantitative evidence for correlations between the NIR absolute magnitudes and J-band light-curve shapes, and demonstrate their utility for distance estimation. The average residual in the Hubble diagram for the training set SNe at cz > 2000kms{sup -1} is 0.10 mag. The new application of bootstrap cross-validation to SN Ia light-curve inference tests the sensitivity of the statistical model fit to the finite sample and estimates the prediction error at 0.15 mag. These results demonstrate that SN Ia NIR light curves are as effective as corrected optical light curves, and, because they are less vulnerable to dust absorption, they have great potential as precise and accurate cosmological distance indicators.

79 ASTRONOMY AND ASTROPHYSICS↗

Nanoporous gold: a hierarchical and multiscale 3D test pattern for characterizing X-ray nano-tomography systems

Full-field transmission X-ray microscopy (TXM) is a well established technique, available at various synchrotron beamlines around the world as well as by laboratory benchtop devices. One of the major TXM challenges, due to its nanometre-scale resolution, is the overall instrument stability during the acquisition of the series of tomographic projections. The ability to correct for vertical and horizontal distortions of each projection image during acquisition is necessary in order to achieve the effective 3D spatial resolution. The effectiveness of such an image alignment is also heavily influenced by the absorption properties and strong contrast of specific features in the scanned sample. Here it is shown that nanoporous gold (NPG) can be used as an ideal 3D test pattern for evaluating and optimizing the performance of a TXM instrument for hard X-rays at a synchrotron beamline. Unique features of NPG, such as hierarchical structures at multiple length scales and high absorbing capabilities, makes it an ideal choice for characterization, which involves a combination of a rapid-alignment algorithm applied on the acquired projections followed by the extraction of a set of both 2D- and 3D-descriptive image parameters. This protocol can be used for comparing the efficiency of TXM instruments at different synchrotron beamlines in the world or benchtop devices, based on a reference library of scanned NPG samples, containing information about the estimated horizontal and vertical alignment values, 2D qualitative parameters and quantitative 3D parameters. The possibility to tailor the ligament sizes of NPG to match the achievable resolution in combination with the high electron density of gold makes NPG an ideal 3D test pattern for evaluating the status and performance of a given synchrotron-based or benchtop-based TXM setup.

36 MATERIALS SCIENCE↗

pGraph: Efficient Parallel Construction of Large-Scale Protein Sequence Homology Graphs

Detecting sequence homology between protein sequences is a fundamental problem in computational molecular biology, with a pervasive application in nearly all analyses that aim to structurally and functionally characterize protein molecules. While detecting the homology between two protein sequences is relatively inexpensive, detecting pairwise homology for a large number of protein sequences can become computationally prohibitive for modern inputs, often requiring millions of CPU hours. Yet, there is currently no robust support to parallelize this kernel. In this paper, we identify the key characteristics that make this problemparticularly hard to parallelize, and then propose a new parallel algorithm that is suited for detecting homology on large data sets using distributed memory parallel computers. Our method, called pGraph, is a novel hybrid between the hierarchical multiple-master/worker model and producer-consumer model, and is designed to break the irregularities imposed by alignment computation and work generation. Experimental results show that pGraph achieves linear scaling on a 2,048 processor distributed memory cluster for a wide range of inputs ranging from as small as 20,000 sequences to 2,560,000 sequences. In addition to demonstrating strong scaling, we present an extensive report on the performance of the various system components and related parametric studies.

Environmental Molecular Sciences Laboratory↗

Total Ozone Mapping Spectrometer (TOMS) Level-3 Data Products User's Guide

Data from the TOMS series of instruments span the time period from November 1978, through the present with about a one and a-half year gap from January 1994 through July 1996. A set of four parameters derived from the TOMS measurements have been archived in the form of daily global maps or Level-3 data products. These products are total column ozone, effective surface reflectivity, aerosol index, and erythermal ultraviolet estimated at the Earth surface. A common fixed grid of I degree latitude by 1.25 degree longitude cells over the entire globe is provided daily for each parameter. These data are archived at the Goddard Space Flight Center Distributed Active Archive Center (DAAQ in Hierarchical Data Format (HDF). They are also available in a character format through the TOMS web site at http://toms.gsfc.nasa.gov. The derivations of the parameters, the mapping algorithm, and the data formats are described. The trend uncertainty for individual TOMS instruments is about 1% decade, but additional uncertainty exists in the combined data record due to uncertainty in the relative calibrations of the various TOMS.

McPeters, Richard D.↗

GENOA-PFA: Progressive Fracture in Composites Simulated Computationally

GENOA-PFA is a commercial version of the Composite Durability Structural Analysis (CODSTRAN) computer program that simulates the progression of damage ultimately leading to fracture in polymer-matrix-composite (PMC) material structures under various loading and environmental conditions. GENOA-PFA offers several capabilities not available in other programs developed for this purpose, making it preferable for use in analyzing the durability and damage tolerance of complex PMC structures in which the fiber reinforcements occur in two- and three-dimensional weaves and braids. GENOA-PFA implements a progressive-fracture methodology based on the idea that a structure fails when flaws that may initially be small (even microscopic) grow and/or coalesce to a critical dimension where the structure no longer has an adequate safety margin to avoid catastrophic global fracture. Damage is considered to progress through five stages: (1) initiation, (2) growth, (3) accumulation (coalescence of propagating flaws), (4) stable propagation (up to the critical dimension), and (5) unstable or very rapid propagation (beyond the critical dimension) to catastrophic failure. The computational simulation of progressive failure involves formal procedures for identifying the five different stages of damage and for relating the amount of damage at each stage to the overall behavior of the deteriorating structure. In GENOA-PFA, mathematical modeling of the composite physical behavior involves an integration of simulations at multiple, hierarchical scales ranging from the macroscopic (lamina, laminate, and structure) to the microscopic (fiber, matrix, and fiber/matrix interface), as shown in the figure. The code includes algorithms to simulate the progression of damage from various source defects, including (1) through-the-thickness cracks and (2) voids with edge, pocket, internal, or mixed-mode delaminations.

Murthy, Pappu L. N.↗

Object detection approach using generative sparse, hierarchical networks with top-down and lateral connections for combining texture/color detection and shape/contour detection

An approach to detecting objects in an image dataset may combine texture/color detection, shape/contour detection, and/or motion detection using sparse, generative, hierarchical models with lateral and top-down connections. A first independent representation of objects in an image dataset may be produced using a color/texture detection algorithm. A second independent representation of objects in the image dataset may be produced using a shape/contour detection algorithm. A third independent representation of objects in the image dataset may be produced using a motion detection algorithm. The first, second, and third independent representations may then be combined into a single coherent output using a combinatorial algorithm.

97 MATHEMATICS AND COMPUTING↗

Object detection approach using generative sparse, hierarchical networks with top-down and lateral connections for combining texture/color detection and shape/contour detection

An approach to detecting objects in an image dataset may combine texture/color detection, shape/contour detection, and/or motion detection using sparse, generative, hierarchical models with lateral and top-down connections. A first independent representation of objects in an image dataset may be produced using a color/texture detection algorithm. A second independent representation of objects in the image dataset may be produced using a shape/contour detection algorithm. A third independent representation of objects in the image dataset may be produced using a motion detection algorithm. The first, second, and third independent representations may then be combined into a single coherent output using a combinatorial algorithm.

97 MATHEMATICS AND COMPUTING↗

Machine Learning Augmented Predictive and Generative Model for Rupture Life in Ferritic and Austenitic Steels

The Larson-Miller parameter (LMP) offers an efficient and fast scheme to estimate the creep rupture life of alloy materials for high temperature applications. However, owing to poor generalizability and dependence on the constant C, which is typically not known a-priori, estimations using the Larson-Miller parameter often result in suboptimal performance for a wide range of materials. At best it is useful in comparing alloys of similar composition. In this work, three machine learning (ML) schemes were developed for rupture life prediction for 9-12% Cr ferritic-martensitic steels and austenitic stainless steels, i.e., a hierarchical model to parameterize LMP using the LMP constant C to compute rupture life, a hierarchical model to parameterize both C and LMP to compute rupture life, and a direct prediction of rupture life. Specifically, we show that the third scheme, using a gradient boosting algorithm, can be used to train ML models for very accurate prediction of rupture life in a variety of alloys (Pear-son Correlation Coefficient > 0.9 for 9-12% Cr and > 0.8 for austenitic stainless steels). In addition, the Shapley value was used to quantify feature importance, making the model interpretable by identifying the effect of various features on the model performance. Furthermore, a variational autoencoder-based generative model was built by conditioning on the experimental dataset to sample hypothetical synthetic candidate alloys from the learnt joint distribution not existing in both 9-12 % Cr ferritic-martensitic steel and austenitic stainless steel datasets. Finally, based on the predictive and generative model, a reinforcement learning strategy has been proposed to guide experimentalists into designing better heat resistant alloys.

Mamun, Md Osman G.↗

Improving ICARUS Track Reconstruction Algorithms

The ICARUS experiment is part of the Short-Baseline Neutrino (SBN) program at Fermilab. The main goal of the experiment is to investigate the possibility of sterile neutrinos in the O(1 eV) mass region and provide clarification of the anomaly detected from the Liquid Scintillator Neutrino Detector (LSND) and MiniBooNE experiments. The ICARUS-T600 detector is a Liquid Argon Time Projection Chamber (LAr-TPC), that can provide excellent 3D imaging and calorimetric reconstruction of any ionizing particles. This detection technique allows a detailed study of neutrino interactions, spanning a wide energy spectrum (from a few keV to several hundreds of GeV). The detector consists of two identical adjacent modules, filled with a total of 760 tons of ultra-pure liquid argon. Each module houses two LAr-TPCs separated by a common cathode with a maximum drift distance of 1.5 m, equivalent to about 1 ms drift time for the nominal $500$ V/m electric drift field. The anode is made of three parallel wire planes positioned 3 mm apart, where the stainless-steel wires are oriented on each plane at a different angle with respect to the horizontal direction ($+60^\degree$,$-60^\degree$,$0^\degree$). The first two planes (Induction 1 and Induction 2) provide a non-destructive charge measurement, whereas the ionization charge is fully collected by the last collection plane. In total, 53248 wires with a 3 mm pitch and length up to 9 m are installed in the detector. In the first stage of the reconstruction, segments of waveforms corresponding to physical signals (hits) are searched for in the deconvolved wire waveform with a threshold-based hit-finding algorithm. Each hit is then fitted with a Gaussian, whose area is proportional to the number of drift electrons generating the signal. In the second stage of the reconstruction, hits are passed as input to Pandora, a framework software composed of different pattern recognition algorithms, that performs a 3D reconstruction of the full image recorded in the collected event, including the identification of interaction vertices and tracks and showers inside the TPC. These are organized into a hierarchical structure (called slice) of particles generated starting from a primary interaction vertex. In some cases, related to the inefficiencies in the hit detection or excessive deflection of the particle trajectory, Pandora breaks the particle's track into two or more smaller pieces and considers each piece as an independent track. We studied this phenomenon focusing on primary muons from ν_μ CC interactions contained in a single module with a track at least 20 cm long, to exclude delta rays. The study determined that about $7-8\%$ of the muon tracks are broken. Approximately $80\%$ of the times, Pandora assigns all segments of the track to the same slice (intra-slice track split), while in the remaining $20\%$ of the cases, one of the segments is associated with another slice (extra-slice track split). To mitigate this phenomenon, we designed an algorithm that detects and stitches the tracks broken by Pandora for the intra-slice split. In Monte Carlo simulations, the algorithm showed an efficiency exceeding $80\%$ and a purity exceeding $93\%$.

Ricci, Alessandro Maria [Pisa U.; INFN, Pisa] (ORC↗

Complex optimization for big computational and experimental neutron datasets

Here, we present a framework to use high performance computing to determine accurate solutions to the inverse optimization problem of big experimental data against computational models. We demonstrate how image processing, mathematical regularization, and hierarchical modeling can be used to solve complex optimization problems on big data. We also demonstrate how both model and data information can be used to further increase solution accuracy of optimization by providing confidence regions for the processing and regularization algorithms. Finally, we use the framework in conjunction with the software package SIMPHONIES to analyze results from neutron scattering experiments on silicon single crystals, and refine first principles calculations to better describe the experimental data.

36 MATERIALS SCIENCE↗

Final report for LDRD project 11-0029 : high-interest event detection in large-scale multi-modal data sets : proof of concept.

Events of interest to data analysts are sometimes difficult to characterize in detail. Rather, they consist of anomalies, events that are unpredicted, unusual, or otherwise incongruent. The purpose of this LDRD was to test the hypothesis that a biologically-inspired anomaly detection algorithm could be used to detect contextual, multi-modal anomalies. There currently is no other solution to this problem, but the existence of a solution would have a great national security impact. The technical focus of this research was the application of a brain-emulating cognition and control architecture (BECCA) to the problem of anomaly detection. One aspect of BECCA in particular was discovered to be critical to improved anomaly detection capabilities: it's feature creator. During the course of this project the feature creator was developed and tested against multiple data types. Development direction was drawn from psychological and neurophysiological measurements. Major technical achievements include the creation of hierarchical feature sets created from both audio and imagery data.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

PHOTOMETRIC PROPERTIES AND LUMINOSITY FUNCTION OF NEARBY MASSIVE EARLY-TYPE GALAXIES

We perform photometric analyses of a bright early-type galaxy sample with 2949 galaxies (M{sub r} < -22.5 mag) in the redshift range of 0.05-0.15, drawn from the Sloan Digital Sky Survey (SDSS) DR7 with morphological classification from Galaxy Zoo 1. We measure the Petrosian and isophotal magnitudes, as well as the corresponding half-light radius for each galaxy. We find that for the brightest galaxies (M{sub r} < -23 mag), our Petrosian magnitudes and isophotal magnitudes to 25 mag arcsec{sup -2} and 1% of the sky brightness are on average 0.16 mag, 0.20 mag, and 0.26 mag brighter than the SDSS Petrosian values, respectively. In the first case, the underestimations are caused by overestimations in the sky background by the SDSS PHOTO algorithm, while the latter two are also due to deeper photometry. Similarly, the typical half-light radii (r{sub 50}) measured by the SDSS algorithm are smaller than our measurements. As a result, the bright end of the r-band luminosity function is found to decline more slowly than previous works. Our measured luminosity densities at the bright end are more than one order of magnitude higher than those of Blanton et al., and the stellar mass densities at M{sub *} {approx} 5 Multiplication-Sign 10{sup 11} M{sub Sun} and M{sub *} {approx} 10{sup 12} M{sub Sun} are a few tenths and a factor of a few higher than those of Bernardi et al. These results may significantly alleviate the tension in the assembly of massive galaxies between observations and predictions of the hierarchical structure formation model.

79 ASTRONOMY AND ASTROPHYSICS↗

Bond order predictions using deep neural networks

Machine learning is an extremely powerful tool for the modern theoretical chemist since it provides a method for bypassing costly algorithms for solving the Schrödinger equation. Already, it has proven able to infer molecular and atomic properties such as charges, enthalpies, dipoles, excited state energies, and others. Most of these machine learning algorithms proceed by inferring properties of individual atoms, even breaking down total molecular energy into individual atomic contributions. In this paper, we introduce a modified version of the Hierarchically Interacting Particle Neural Network (HIP-NN) capable of making predictions on the bonds between atoms rather than on the atoms themselves. We train the modified HIP-NN to infer bond orders for a large number of small organic molecules as computed via the Natural Bond Orbital package. We demonstrate that the trained model is extensible to molecules much larger than those in the training set by studying its performance on the COMP6 dataset. This method has applications in cheminformatics and force field parameterization and opens a promising future for machine learning models to predict other quantities that are defined between atoms such as density matrix elements, Hamiltonian parameters, and molecular reactivities.

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↗