Search NASA⌕ Search

SEARCH · Search NASA

Results for “Kernel 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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

SU-F-T-428: An Optimization-Based Commissioning Tool for Finite Size Pencil Beam Dose Calculations

Purpose: Finite size pencil beam (FSPB) algorithms are commonly used to pre-calculate the beamlet dose distribution for IMRT treatment planning. FSPB commissioning, which usually requires fine tuning of the FSPB kernel parameters, is crucial to the dose calculation accuracy and hence the plan quality. Yet due to the large number of beamlets, FSPB commissioning could be very tedious. This abstract reports an optimization-based FSPB commissioning tool we have developed in MatLab to facilitate the commissioning. Methods: A FSPB dose kernel generally contains two types of parameters: the profile parameters determining the dose kernel shape, and a 2D scaling factors accounting for the longitudinal and off-axis corrections. The former were fitted using the penumbra of a reference broad beam’s dose profile with Levenberg-Marquardt algorithm. Since the dose distribution of a broad beam is simply a linear superposition of the dose kernel of each beamlet calculated with the fitted profile parameters and scaled using the scaling factors, these factors could be determined by solving an optimization problem which minimizes the discrepancies between the calculated dose of broad beams and the reference dose. Results: We have commissioned a FSPB algorithm for three linac photon beams (6MV, 15MV and 6MVFFF). Dose of four field sizes (6*6cm2, 10*10cm2, 15*15cm2 and 20*20cm2) were calculated and compared with the reference dose exported from Eclipse TPS system. For depth dose curves, the differences are less than 1% of maximum dose after maximum dose depth for most cases. For lateral dose profiles, the differences are less than 2% of central dose at inner-beam regions. The differences of the output factors are within 1% for all the three beams. Conclusion: We have developed an optimization-based commissioning tool for FSPB algorithms to facilitate the commissioning, providing sufficient accuracy of beamlet dose calculation for IMRT optimization.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2015

Three-dimensional Fréchet sensitivity kernels for electromagnetic wave propagation

Electromagnetic imaging methods are useful tools for monitoring subsurface changes in pore-fluid content and the associated changes in electrical permittivity and conductivity. The most common method for georadar tomography uses a high frequency ray-theoretic approximation that is valid when material variations are sufficiently small relative to the wavelength of the propagating wave. Georadar methods, however, often utilize electromagnetic waves that propagate within heterogeneous media at frequencies where ray theory may not be applicable. In this paper we describe the 3-D Fréchet sensitivity kernels for EM wave propagation. Various data functional types are formulated that consider all three components of the electric wavefield and incorporate near-, intermediate-, and far-field contributions. We show that EM waves exhibit substantial variations for different relative source-receiver component orientations. The 3-D sensitivities also illustrate out-of-plane effects that are not captured in 2-D sensitivity kernels and can influence results obtained using 2-D inversion methods to image structures that are in reality 3-D.

58 GEOSCIENCES↗
DOE OSTI2021

A New Class of AMG Interpolation Methods Based on Matrix-Matrix Multiplications

A new class of distance-two interpolation methods for algebraic multigrid (AMG) that can be formulated in terms of sparse matrix-matrix multiplications is presented and analyzed. Compared with similar distance-two prolongation operators, the proposed algorithms exhibit improved efficiency and portability to various computing platforms, since they allow one to easily exploit existing high-performance sparse matrix kernels. The new interpolation methods have been implemented in hypre, a widely used parallel multigrid solver library. With the proposed interpolations, the overall time of hypre's BoomerAMG setup can be considerably reduced, while sustaining equivalent, sometimes improved, convergence rates. Numerical results for a variety of test problems on parallel machines are presented that support the superiority of the proposed interpolation operators over the existing ones in hypre.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Functional differentiability in time-dependent quantum mechanics

In this work, we investigate the functional differentiability of the time-dependent many-body wave function and of derived quantities with respect to time-dependent potentials. For properly chosen Banach spaces of potentials and wave functions, Fréchet differentiability is proven. From this follows an estimate for the difference of two solutions to the time-dependent Schrödinger equation that evolve under the influence of different potentials. Such results can be applied directly to the one-particle density and to bounded operators, and present a rigorous formulation of non-equilibrium linear-response theory where the usual Lehmann representation of the linear-response kernel is not valid. Further, the Fréchet differentiability of the wave function provides a new route towards proving basic properties of time-dependent density-functional theory.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2014

A Framework for Adaptable Operating and Runtime Systems

The emergence of new classes of HPC systems where performance improvement is enabled by Moore’s Law for technology is manifest through multi-core-based architectures including specialized GPU structures. Operating systems were originally designed for control of uniprocessor systems. By the 1980s multiprogramming, virtual memory, and network interconnection were integral services incorporated as part of most modern computers. HPC operating systems were primarily derivatives of the Unix model with Linux dominating the Top-500 list. The use of Linux for commodity clusters was first pioneered by the NASA Beowulf Project. However, the rapid increase in number of cores to achieve performance gain through technology advances has exposed the limitations of POSIX general-purpose operating systems in scaling and efficiency. This project was undertaken through the leadership of Sandia National Laboratories and in partnership of the University of New Mexico to investigate the alternative of composable lightweight kernels on scalable HPC architectures to achieve superior performance for a wide range of applications. The use of composable operating systems is intended to provide a minimalist set of services specifically required by a given application to preclude overheads and operational uncertainties (“OS noise”) that have been demonstrated to degrade efficiency and operational consistency. This project was undertaken as an exploration to investigate possible strategies and methods for composable lightweight kernel operating systems towards support for extreme scale systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

In situ multi-tier auto-ignition detection applied to dual-fuel combustion simulations

Here we use an anomaly detection methodology that is centered on analyzing fourth-order joint moments (co-kurtosis), particularly focusing on its application in auto-ignition of combustion problems with large numbers of species. Unsupervised anomaly detection is challenging to generalize across problem types and domains. A recent technique, centered on analyzing information in the fourth-order joint moment co-kurtosis, has shown promise, especially for high-dimensional scientific data. In this work we present developments to the co-kurtosis based anomaly detection method needed to make it effective and scalable for large-scale distributed scientific data, such as those generated by massively parallel simulations. An in situ co-kurtosis algorithm is employed as the anomaly detection method for identifying ignition kernels in simulations of turbulent combustion. Here, we extend an existing methodology which identifies regions of the domain where anomalies are present, and add another tier of anomaly detection where the individual samples contributing to the anomaly are identified. We apply this algorithm on-the-fly to a variety of turbulent reacting flow problems and compare it to the widely used (but significantly more expensive) chemical explosive mode analysis (CEMA). We demonstrate the ability of the method to detect and identify the onset of low and high temperature ignition which can be used for computational steering, as chemical and combustion anomalies occur intermittently at spatio-temporal locations unknown a priori. Finally, we apply our lightweight in situ algorithm to an exascale high-fidelity simulation with a total of 2.4 Trillion degrees of freedom, performed using an adaptive mesh refinement solver. Furthermore, through a scalability analysis, we show that the relative computational cost of this in-situ anomaly detection algorithm compared to an iteration of the reacting flow solver is negligible.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Highly Efficient Compensation-Based Parallelism for Wavefront Loops on GPUs

Wavefront loops are widely used in many scientific applications, e.g., partial differential equation (PDE) solvers and sequence alignment tools. However, due to the data dependencies in wavefront loops, it is challenging to fully utilize the abundant compute units of GPUs and to reuse data through their memory hierarchy. Existing solutions can only optimize for these factors to a limited extent. For example, tiling-based methods optimize memory access but may result in load imbalance; while compensation-based methods, which change the original order of computation to expose more parallelism and then compensate for it, suffer from both global synchronization overhead and limited generality. In this paper, we first prove under which circumstances that breaking data dependencies and properly changing the sequence of computation operators in our compensation-based method does not affect the correctness of results. Based on this analysis, we design a highly efficient compensation-based parallelism on GPUs. Our method provides weighted scan-based GPU kernels to optimize the computation and combines with the tiling method to optimize memory access and synchronization. The performance results on the NVIDIA K80 and P100 GPU platforms demonstrate that our method can achieve significant improvements for four types of real-world application kernels over the state-of-the-art research.

Hou, Kaixi↗
DOE OSTI2013

Flux and Reaction Rate Kernel Density Estimators in OpenMC

Current methods for obtaining flux and reaction rate profiles in a Monte Carlo simulation rely on histogram tallies that can suffer from large uncertainties when fine detail is required. Recently, another density estimation technique has been applied to Monte Carlo radiation transport sim ulations: the Kernel Density Estimator (KDE) [Banerjee, K.]. KDEs are capable of calculating the flux at a point without the necessity of tracking particles on a spatial mesh, making KDEs ideal tools for multi-physics problems. Furthermore, KDEs allow for one collision or track length to contribute to the estimate of the flux at multiple points, circumventing the large statistical uncertainties associated with fine-mesh histogram tallies. However, KDEs have only been proven to be capable of estimating flux values at a series of points; no work has been done to estimate reaction rates or other quantities of interest. Furthermore, little work has been done in optimizing the KDE bandwidth for the energy-dependence of neutron transport simulations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2015

Optical properties of alkali halide crystals from all-electron hybrid TD-DFT calculations

We present a study of the electronic and optical properties of a series of alkali halide crystals AX, with A = Li, Na, K, Rb and X = F, Cl, Br based on a recent implementation of hybrid-exchange time-dependent density functional theory (TD-DFT) (TD-B3LYP) in the all-electron Gaussian basis set code CRYSTAL. We examine, in particular, the impact of basis set size and quality on the prediction of the optical gap and exciton binding energy. The formation of bound excitons by photoexcitation is observed in all the studied systems and this is shown to be correlated to specific features of the Hartree-Fock exchange component of the TD-DFT response kernel. All computed optical gaps and exciton binding energies are however markedly below estimated experimental and, where available, 2-particle Green’s function (GW-Bethe-Salpeter equation, GW-BSE) values. We attribute this reduced exciton binding to the incorrect asymptotics of the B3LYP exchange correlation ground state functional and of the TD-B3LYP response kernel, which lead to a large underestimation of the Coulomb interaction between the excited electron and hole wavefunctions. Considering LiF as an example, we correlate the asymptotic behaviour of the TD-B3LYP kernel to the fraction of Fock exchange admixed in the ground state functional c{sub HF} and show that there exists one value of c{sub HF} (∼0.32) that reproduces at least semi-quantitatively the optical gap of this material.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2010

Shading correction for on-board cone-beam CT in radiation therapy using planning MDCT images

Purpose: Applications of cone-beam CT (CBCT) to image-guided radiation therapy (IGRT) are hampered by shading artifacts in the reconstructed images. These artifacts are mainly due to scatter contamination in the projections but also can result from uncorrected beam hardening effects as well as nonlinearities in responses of the amorphous silicon flat panel detectors. While currently, CBCT is mainly used to provide patient geometry information for treatment setup, more demanding applications requiring high-quality CBCT images are under investigation. To tackle these challenges, many CBCT correction algorithms have been proposed; yet, a standard approach still remains unclear. In this work, we propose a shading correction method for CBCT that addresses artifacts from low-frequency projection errors. The method is consistent with the current workflow of radiation therapy. Methods: With much smaller inherent scatter signals and more accurate detectors, diagnostic multidetector CT (MDCT) provides high quality CT images that are routinely used for radiation treatment planning. Using the MDCT image as ''free'' prior information, we first estimate the primary projections in the CBCT scan via forward projection of the spatially registered MDCT data. Since most of the CBCT shading artifacts stem from low-frequency errors in the projections such as scatter, these errors can be accurately estimated by low-pass filtering the difference between the estimated and raw CBCT projections. The error estimates are then subtracted from the raw CBCT projections. Our method is distinct from other published correction methods that use the MDCT image as a prior because it is projection-based and uses limited patient anatomical information from the MDCT image. The merit of CBCT-based treatment monitoring is therefore retained. Results: The proposed method is evaluated using two phantom studies on tabletop systems. On the Catphan(c)600 phantom, our approach reduces the reconstruction error from 348 Hounsfield unit (HU) without correction to 4 HU around the object center after correction, and from 375 HU to 17 HU in the high-contrast regions. In the selected regions of interest (ROIs), the average image contrast is increased by a factor of 3.3. When noise suppression is implemented, the proposed correction substantially improves the contrast-to-noise ratio (CNR) and therefore the visibility of low-contrast objects, as seen in a more challenging pelvis phantom study. Besides a significant improvement in image uniformity, a low-contrast object of {approx}25 HU, which is otherwise buried in the shading artifacts, can be clearly identified after the proposed correction due to a CNR increase of 3.1. Compared to a kernel-based scatter correction method coupled with an analytical beam hardening correction, our approach also shows an overall improved performance with some residual artifacts. Conclusions: By providing effective shading correction, our approach has the potential to improve the accuracy of more advanced CBCT-based clinical applications for IGRT, such as tumor delineation and dose calculation.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Parallel hybrid quantum-classical machine learning for kernelized time-series classification

Supervised time-series classification garners widespread interest because of its applicability throughout a broad application domain including finance, astronomy, biosensors, and many others. Here, in this work, we tackle this problem with hybrid quantum-classical machine learning, deducing pairwise temporal relationships between time-series instances using a timeseries Hamiltonian kernel (TSHK). A TSHK is constructed with a sum of inner products generated by quantum states evolved using a parameterized time evolution operator. This sum is then optimally weighted using techniques derived from multiple kernel learning. Because we treat the kernel weighting step as a differentiable convex optimization problem, our method can be regarded as an end-to-end learnable hybrid quantum-classical-convex neural network, or QCC-net, whose output is a data set-generalized kernel function suitable for use in any kernelized machine learning technique such as the support vector machine (SVM). Using our TSHK as input to a SVM, we classify univariate and multivariate time-series using quantum circuit simulators and demonstrate the efficient parallel deployment of the algorithm to 127-qubit superconducting quantum processors using quantum multi-programming.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Nonparametric, data-based kernel interpolation for particle-tracking simulations and kernel density estimation

Traditional interpolation techniques for particle tracking include binning and convolutional formulas that use pre-determined (i.e., closed-form, parameteric) kernels. In many instances, the particles are introduced as point sources in time and space, so the cloud of particles (either in space or time) is a discrete representation of the Green’s function of an underlying PDE. As such, each particle is a sample from the Green’s function; therefore, each particle should be distributed according to the Green’s function. In short, the kernel of a convolutional interpolation of the particle sample “cloud” should be a replica of the cloud itself. This idea gives rise to an iterative method by which the form of the kernel may be discerned in the process of interpolating the Green’s function. When the Green’s function is a density, this method is broadly applicable to interpolating a kernel density estimate based on random data drawn from a single distribution. We formulate and construct the algorithm and demonstrate its ability to perform kernel density estimation of skewed and/or heavy-tailed data including breakthrough curves.

42 ENGINEERING↗
DOE OSTI2011

Development and testing of improved statistical wind power forecasting methods.

Wind power forecasting (WPF) provides important inputs to power system operators and electricity market participants. It is therefore not surprising that WPF has attracted increasing interest within the electric power industry. In this report, we document our research on improving statistical WPF algorithms for point, uncertainty, and ramp forecasting. Below, we provide a brief introduction to the research presented in the following chapters. For a detailed overview of the state-of-the-art in wind power forecasting, we refer to [1]. Our related work on the application of WPF in operational decisions is documented in [2]. Point forecasts of wind power are highly dependent on the training criteria used in the statistical algorithms that are used to convert weather forecasts and observational data to a power forecast. In Chapter 2, we explore the application of information theoretic learning (ITL) as opposed to the classical minimum square error (MSE) criterion for point forecasting. In contrast to the MSE criterion, ITL criteria do not assume a Gaussian distribution of the forecasting errors. We investigate to what extent ITL criteria yield better results. In addition, we analyze time-adaptive training algorithms and how they enable WPF algorithms to cope with non-stationary data and, thus, to adapt to new situations without requiring additional offline training of the model. We test the new point forecasting algorithms on two wind farms located in the U.S. Midwest. Although there have been advancements in deterministic WPF, a single-valued forecast cannot provide information on the dispersion of observations around the predicted value. We argue that it is essential to generate, together with (or as an alternative to) point forecasts, a representation of the wind power uncertainty. Wind power uncertainty representation can take the form of probabilistic forecasts (e.g., probability density function, quantiles), risk indices (e.g., prediction risk index) or scenarios (with spatial and/or temporal dependence). Statistical approaches to uncertainty forecasting basically consist of estimating the uncertainty based on observed forecasting errors. Quantile regression (QR) is currently a commonly used approach in uncertainty forecasting. In Chapter 3, we propose new statistical approaches to the uncertainty estimation problem by employing kernel density forecast (KDF) methods. We use two estimators in both offline and time-adaptive modes, namely, the Nadaraya-Watson (NW) and Quantilecopula (QC) estimators. We conduct detailed tests of the new approaches using QR as a benchmark. One of the major issues in wind power generation are sudden and large changes of wind power output over a short period of time, namely ramping events. In Chapter 4, we perform a comparative study of existing definitions and methodologies for ramp forecasting. We also introduce a new probabilistic method for ramp event detection. The method starts with a stochastic algorithm that generates wind power scenarios, which are passed through a high-pass filter for ramp detection and estimation of the likelihood of ramp events to happen. The report is organized as follows: Chapter 2 presents the results of the application of ITL training criteria to deterministic WPF; Chapter 3 reports the study on probabilistic WPF, including new contributions to wind power uncertainty forecasting; Chapter 4 presents a new method to predict and visualize ramp events, comparing it with state-of-the-art methodologies; Chapter 5 briefly summarizes the main findings and contributions of this report.

17 WIND ENERGY↗
DOE OSTI2022

Active learning of reactive Bayesian force fields applied to heterogeneous catalysis dynamics of H/Pt

Abstract Atomistic modeling of chemically reactive systems has so far relied on either expensive ab initio methods or bond-order force fields requiring arduous parametrization. Here, we describe a Bayesian active learning framework for autonomous “on-the-fly” training of fast and accurate reactive many-body force fields during molecular dynamics simulations. At each time-step, predictive uncertainties of a sparse Gaussian process are evaluated to automatically determine whether additional ab initio training data are needed. We introduce a general method for mapping trained kernel models onto equivalent polynomial models whose prediction cost is much lower and independent of the training set size. As a demonstration, we perform direct two-phase simulations of heterogeneous H 2 turnover on the Pt(111) catalyst surface at chemical accuracy. The model trains itself in three days and performs at twice the speed of a ReaxFF model, while maintaining much higher fidelity to DFT and excellent agreement with experiment.

42 ENGINEERING↗
DOE OSTI2016

Field-gradient partitioning for fracture and frictional contact in the material point method: Field-gradient partitioning for fracture and frictional contact in the material point method [Fracture and frictional contact in material point method using damage-field gradients for velocity-field partitioning]

Contact and fracture in the material point method require grid-scale enrichment or partitioning of material into distinct velocity fields to allow for displacement or velocity discontinuities at a material interface. We present a new method which a kernel-based damage field is constructed from the particle data. The gradient of this field is used to dynamically repartition the material into contact pairs at each node. Our approach avoids the need to construct and evolve explicit cracks or contact surfaces and is therefore well suited to problems involving complex 3-D fracture with crack branching and coalescence. A straightforward extension of this approach permits frictional ‘self-contact’ between surfaces that are initially part of a single velocity field, enabling more accurate simulation of granular flow, porous compaction, fragmentation, and comminution of brittle materials. Finally, numerical simulations of self contact and dynamic crack propagation are presented to demonstrate the accuracy of the approach.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2011

Correction for patient table-induced scattered radiation in cone-beam computed tomography (CBCT)

Purpose: In image-guided radiotherapy, an artifact typically seen in axial slices of x-ray cone-beam computed tomography (CBCT) reconstructions is a dark region or ''black hole'' situated below the scan isocenter. The authors trace the cause of the artifact to scattered radiation produced by radiotherapy patient tabletops and show it is linked to the use of the offset-detector acquisition mode to enlarge the imaging field-of-view. The authors present a hybrid scatter kernel superposition (SKS) algorithm to correct for scatter from both the object-of-interest and the tabletop. Methods: Monte Carlo simulations and phantom experiments were first performed to identify the source of the black hole artifact. For correction, a SKS algorithm was developed that uses separate kernels to estimate scatter from the patient tabletop and the object-of-interest. Each projection is divided into two regions, one defined by the shadow cast by the tabletop on the imager and one defined by the unshadowed region. The region not shadowed by the tabletop is processed using the recently developed fast adaptive scatter kernel superposition (fASKS) method which employs asymmetric kernels that best model scatter transport through bodylike objects. The shadowed region is convolved with a combination of slab-derived symmetric SKS kernels and asymmetric fASKS kernels. The composition of the hybrid kernels is projection-angle-dependent. To test the algorithm, pelvis phantom and in vivo data were acquired using a CBCT test stand, a Varian Acuity simulator, and a Varian On-Board Imager, all of which have similar geometries and components. Artifact intensities and Hounsfield unit (HU) accuracies in the reconstructions were assessed before and after the correction. Results: The hybrid kernel algorithm provided effective correction and produced substantially better scatter estimates than the symmetric SKS or asymmetric fASKS methods alone. HU nonuniformities in the reconstructed pelvis phantom were reduced from 220 to 50 HU (i.e., 22%-5%). In the in vivo scans, the black hole artifact was reduced by up to 147 HU, a 73% improvement, and anatomical details in the prostate and rectum areas were made considerably more visible. Conclusions: Radiotherapy tabletops, which are generally flatter and larger than those used for diagnostic CT, can produce significant scatter-related artifacts. The proposed hybrid SKS algorithm accurately estimates scatter from both the object-of-interest and the patient tabletop, and resulting image uniformities and HU accuracies are greatly improved.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

BTZ one-loop determinants via the Selberg zeta function for general spin

We relate the heat kernel and quasinormal mode methods of computing the 1-loop partition function of arbitrary spin fields on a rotating (Euclidean) BTZ background using the Selberg zeta function associated with H 3 /Z, extending (arXiv:1811.08433) [1]. Previously, Perry and Williams [2] showed for a scalar field that the zeros of the Selberg zeta function coincide with the poles of the associated scattering operator upon a relabeling of integers. We extend the integer relabeling to the case of general spin, and discuss its relationship to the removal of non-square-integrable Euclidean zero modes.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

Crack propagation in silica from reactive classical molecular dynamics simulations

Mechanistic insight into the process of crack growth can be obtained through molecular dynamics (MD) simulations. In this investigation of fracture propagation, a slit crack was introduced into an atomistic amorphous silica model and mode I stress was applied through far–field loading until the crack propagates. Atomic displacements and forces and an Irving–Kirkwood method with a Lagrangian kernel estimator were used to calculate the J –integral of classical fracture mechanics around the crack tip. The resulting fracture toughness (K IC ), 0.76 ± 0.16 MPa√m, agrees with experimental values. In addition, the stress fields and dissipation energies around the slit crack indicate the development of an inelastic region ~30Å in diameter. This is one of the first reports of KIC values obtained from up–scaled atomic–level energies and stresses through the J –integral. In conclusion, the application of the ReaxFF classical MD force field in this study provides the basis for future research into crack growth in multicomponent oxides in a variety of environmental conditions.

36 MATERIALS SCIENCE↗