Search NASA⌕ Search

SEARCH · Search NASA

Results for “Convolution”

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 775 records · Page 43

First-Principles Treatment of Vibrational Broadening in X-Ray Excited Valence Band Spectra for n-SrTiO 3 (001)

The valence band maximum (VBM) is an important quantity for semiconductors as it locates the Fermi level relative to the band edge. Accurate measurement of this quantity in near-surface regions of semiconductors by photoemission is a first step toward determining the electronic properties of heterostructures involving these materials. While extrapolating the leading edge of the valence band to the energy axis in photoemission spectra is a widely used way to find the VBM, this method can be ambiguous if the leading edge exhibits multiple slopes. Another way to determine the VBM is to fit the leading edge to an appropriately broadened, cross-section modulated theoretical density of states (DOS). Three kinds of broadening that should be included for maximum accuracy are those due to: (1) finite instrumental resolution, (2) valence hole lifetime, and (3) vibrational excitations. While steps (1) and (2) are straightforward to implement, (3) is more difficult because the appropriate amount of broadening is not known a priori. Here, we demonstrate that explicit inclusion of vibrational broadening using ab initio molecular dynamics facilitates accurate VBM determination for n-SrTiO 3 (001). The total DOS is constructed by summing time-averaged projections at elevated temperature onto s - , p - , and d orbitals for the constituent atoms and modulating with the associated photoemission cross sections. Subsequent convolutions of the total DOS, first with a Gaussian of width equal to the experimental energy resolution and second with a Lorentzian to simulate valence hole lifetime effects, yield line shapes that reproduce the experimental leading edges rather well. The VBM is then given by the energy at which the vibrationally broadened total DOS (prior to the convolutions) goes to zero. The VBMs generated by this method quantitatively agree with those resulting from extrapolating from the middle of the measured leading edge for SrTiO 3 .

36 MATERIALS SCIENCE↗

Spin structure functions of {sup 3}He at finite Q{sup 2}

Using recently derived relations between spin-dependent nuclear and nucleon g{sub 1} and g{sub 2} structure functions at finite Q{sup 2}, we study nuclear effects in {sup 3}He in the nucleon resonance and deep inelastic regions. Comparing the finite-Q{sup 2} results with the standard convolution formulas obtained in the large-Q{sup 2} limit, we find significant broadening of the effective nucleon momentum distribution functions, leading to additional suppression of the nuclear g{sub 1} and g{sub 2} structure functions around the resonance peaks.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Learning to identify electrons

In this report we investigate whether state-of-the-art classification features commonly used to distinguish electrons from jet backgrounds in collider experiments are overlooking valuable information. A deep convolutional neural network analysis of electromagnetic and hadronic calorimeter deposits is compared to the performance of typical features, revealing a ≈ 5% gap which indicates that these lower-level data do contain untapped classification power. To reveal the nature of this unused information, we use a recently developed technique to map the deep network into a space of physically interpretable observables. We identify two simple calorimeter observables which are not typically used for electron identification, but which mimic the decisions of the convolutional network and nearly close the performance gap.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Higher twist parton distributions from light-cone wave functions

We explore the possibility to construct higher-twist parton distributions in a nucleon at some low reference scale from convolution integrals of the light-cone wave functions (WFs). To this end we introduce simple models for the four-particle nucleon WFs involving three valence quarks and a gluon with total orbital momentum zero, and estimate their normalization (WF at the origin) using QCD sum rules. We demonstrate that these WFs provide one with a reasonable description of both polarized and unpolarized parton densities at large values of the Bjorken variable x{>=}0.5. Twist-three parton distributions are then constructed as convolution integrals of qqqg and the usual three-quark WFs. The cases of the polarized structure function g{sub 2}(x,Q{sup 2}) and single transverse spin asymmetries are considered in detail. We find that the so-called gluon pole contribution to twist-three distributions relevant for single spin asymmetry vanishes in this model, but is generated perturbatively at higher scales by the evolution, in the spirit of Glueck-Reya-Vogt parton distributions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Conceptual designs of two petawatt-class pulsed-power accelerators for high-energy-density-physics experiments

Here, we have developed conceptual designs of two petawatt-class pulsed-power accelerators: Z 300 and Z 800. The designs are based on an accelerator architecture that is founded on two concepts: single-stage electrical-pulse compression and impedance matching [Phys. Rev. ST Accel. Beams 10, 030401 (2007)]. The prime power source of each machine consists of 90 linear-transformer-driver (LTD) modules. Each module comprises LTD cavities connected electrically in series, each of which is powered by 5-GW LTD bricks connected electrically in parallel. (A brick comprises a single switch and two capacitors in series.) Six water-insulated radial-transmission-line impedance transformers transport the power generated by the modules to a six-level vacuum-insulator stack. The stack serves as the accelerator’s water-vacuum interface. The stack is connected to six conical outer magnetically insulated vacuum transmission lines (MITLs), which are joined in parallel at a 10-cm radius by a triple-post-hole vacuum convolute. The convolute sums the electrical currents at the outputs of the six outer MITLs, and delivers the combined current to a single short inner MITL. The inner MITL transmits the combined current to the accelerator’s physics-package load. Z 300 is 35 m in diameter and stores 48 MJ of electrical energy in its LTD capacitors. The accelerator generates 320 TW of electrical power at the output of the LTD system, and delivers 48 MA in 154 ns to a magnetized-liner inertial-fusion (MagLIF) target [Phys. Plasmas 17, 056303 (2010)]. The peak electrical power at the MagLIF target is 870 TW, which is the highest power throughout the accelerator. Power amplification is accomplished by the centrally located vacuum section, which serves as an intermediate inductive-energy-storage device. The principal goal of Z 300 is to achieve thermonuclear ignition; i.e., a fusion yield that exceeds the energy transmitted by the accelerator to the liner. 2D magnetohydrodynamic (MHD) simulations suggest Z 300 will deliver 4.3 MJ to the liner, and achieve a yield on the order of 18 MJ. Z 800 is 52 m in diameter and stores 130 MJ. This accelerator generates 890 TW at the output of its LTD system, and delivers 65 MA in 113 ns to a MagLIF target. The peak electrical power at the MagLIF liner is 2500 TW. The principal goal of Z 800 is to achieve high-yield thermonuclear fusion; i.e., a yield that exceeds the energy initially stored by the accelerator’s capacitors. 2D MHD simulations suggest Z 800 will deliver 8.0 MJ to the liner, and achieve a yield on the order of 440 MJ. Z 300 and Z 800, or variations of these accelerators, will allow the international high-energy-density-physics community to conduct advanced inertial-confinement-fusion, radiation-physics, material-physics, and laboratory-astrophysics experiments over heretofore-inaccessible parameter regimes.

42 ENGINEERING↗

Theory for Equivariant Quantum Neural Networks

Quantum neural network architectures that have little to no inductive biases are known to face trainability and generalization issues. Inspired by a similar problem, recent breakthroughs in machine learning address this challenge by creating models encoding the symmetries of the learning task. This is materialized through the usage of equivariant neural networks the action of which commutes with that of the symmetry. In this work, we import these ideas to the quantum realm by presenting a comprehensive theoretical framework to design equivariant quantum neural networks (EQNNs) for essentially any relevant symmetry group. We develop multiple methods to construct equivariant layers for EQNNs and analyze their advantages and drawbacks. Our methods can find unitary or general equivariant quantum channels efficiently even when the symmetry group is exponentially large or continuous. As a special implementation, we show how standard quantum convolutional neural networks (QCNNs) can be generalized to group-equivariant QCNNs where both the convolution and pooling layers are equivariant to the symmetry group. We then numerically demonstrate the effectiveness of a S U ( 2 ) -equivariant QCNN over symmetry-agnostic QCNN on a classification task of phases of matter in the bond-alternating Heisenberg model. Our framework can be readily applied to virtually all areas of quantum machine learning. Lastly, we discuss about how symmetry-informed models such as EQNNs provide hopes to alleviate central challenges such as barren plateaus, poor local minima, and sample complexity. Published by the American Physical Society 2024

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

`Pink'-beam X-ray powder diffraction profile and its use in Rietveld refinement

The powder diffraction profile obtained with a pink-beam synchrotron X-ray source is described as the convolution of a back-to-back pair of exponentials convoluted with the Gaussian and Lorentzian components of a pseudo-Voigt. We report this new function is employed for the first time in a Rietveld refinement using data collected from a single 100 ps synchrotron X-ray micropulse.

36 MATERIALS SCIENCE↗

A comparison of deep-learning-based inpainting techniques for experimental X-ray scattering

The implementation is proposed of image inpainting techniques for the reconstruction of gaps in experimental X-ray scattering data. The proposed methods use deep learning neural network architectures, such as convolutional autoencoders, tunable U-Nets, partial convolution neural networks and mixed-scale dense networks, to reconstruct the missing information in experimental scattering images. In particular, the recovered pixel intensities are evaluated against their corresponding ground-truth values using the mean absolute error and the correlation coefficient metrics. The results demonstrate that the proposed methods achieve better performance than traditional inpainting algorithms such as biharmonic functions. Overall, tunable U-Net and mixed-scale dense network architectures achieved the best reconstruction performance among all the tested algorithms, with correlation coefficient scores greater than 0.9980.

97 MATHEMATICS AND COMPUTING↗

Retrofitting Word Embeddings with the UMLS Metathesaurus for Clinical Information Extraction

Deep learning has surged in popularity and proven to be effective for various artificial intelligence applications including information extraction from cancer pathology reports. Since word representation is a core unit that enables deep learning algorithms to understand words and be able to perform NLP, this representation must include as much information as possible to help these algorithms achieve high classification performance. Therefore, in this work in addition to the distributional information of words in large sized corpora, we use UMLS vocabulary resources to enrich the vector space representation of words with the semantic relations between words. These resources provide many terminologies pertaining to cancer. The refined word embeddings are used with a convolutional neural (CNN) model to extract four data elements from cancer pathology reports; ICD-O-3 tumor topography codes, tumor laterality, behavior, and histological grade. We observed that using UMLS vocabulary resources to enrich word embeddings of CNN models consistently outperformed CNN models without pre-training word embeddings and even with pre-trained word embeddings on a domain specific corpus across all four tasks. The results show marginal improvement on the laterality task, but a significant improvement on the other tasks, especially for the macro-f score. Specifically, the improvements are 3%, 13%, and 15% for tumor site, histological grade, and behavior tasks, respectively. This approach is encouraging to enrich word embeddings with more clinical data resources to be used for information abstraction tasks from clinical pathology reports.

Alawad, Mohammed↗

Deep Learning-Based Dynamic Modeling of Three-Phase Voltage Source Inverters

Inverter-based resource (IBR) models are necessary to analyze modern power system stability and create effective control strategies. Modeling IBRs in converter-rich power systems is crucial, yet challenging due to the lack of commercial information on converter topologies and control parameters. This paper proposes novel convolutional neural network (CNN)–based data-driven techniques for modeling IBRs, addressing adaptability and proprietary concerns without requiring internal system physics knowledge. The proposed method is tested using real grid-tied commercial IBR transient data and demonstrates effectiveness and accuracy. Furthermore, the developed modeling approach is integrated and implemented in the open-source power distribution simulation and analysis tool, GridLAB-D, to illustrate the potentiality of dynamic analysis of large-scale power systems with high IBRs.

deep learning, artificial intelligence↗

JACC.shared: Leveraging HPC Metaprogramming and Performance Portability for Computations That Use Shared Memory GPUs

In this work, we present JACC.shared, a new feature of Julia for ACCelerators (JACC), which is the performanceportable and metaprogramming model of the just-in-time and LLVM-based Julia language. This new feature allows JACC applications to leverage the high-performance computing (HPC) capabilities of high-bandwidth, on-chip GPU memory. Historically, exploiting high-bandwidth, shared-memory GPUs has not been a priority for high-level programming solutions. JACC.shared covers that gap for the first time, thereby providing a highlevel, portable, and easy-to-use solution for programmers to exploit this memory and supporting all current major accelerator architectures. Well-known HPC and AI workloads, such as multi/hyperspectral imaging and AI convolutions, have been used to evaluate JACC.shared on two exascale GPU architectures hosted by some of the most powerful US Department of Energy supercomputers: Perlmutter (NVIDIA A100) and Frontier (AMD MI250X). The performance evaluation reports speedup of up to 3.5× by adding only one line of code to the base codes, thus providing important accelerators in a simple, portable, and transparent way and elevating the programming productivity and performance-portability capabilities for Julia/JACC HPC, AI, and scientific applications.

Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗

Autonomous Human Activity Classification From Wearable Multi-Modal Sensors

There has been significant amount of research work on human activity classification relying either on Inertial Measurement Unit (IMU) data or data from static cameras providing a third-person view. There has been relatively less work using wearable cameras, providing first-person or egocentric view, and even fewer approaches combining egocentric video with IMU data. Utilizing only IMU data limits the variety and complexity of the activities that can be detected. For instance, the sitting activity can be detected by IMU data, but it cannot be determined whether the subject has sat on a chair or a sofa, or where the subject is. To perform fine-grained activity classification, and to distinguish between activities that cannot be differentiated by only IMU data, we present an autonomous and robust method using data from both wearable cameras and IMUs. In contrast to convolutional neural network-based approaches, we propose to employ capsule networks to obtain features from egocentric video data. Moreover, Convolutional Long Short Term Memory framework is employed both on egocentric videos and IMU data to capture the temporal aspect of actions. We also propose a genetic algorithm-based approach to autonomously and systematically set various network parameters, rather than using manual settings. Experiments have been conducted to perform 9- and 26-label activity classification, and the proposed method, using autonomously set network parameters, has provided very promising results, achieving overall accuracies of 86.6% and 77.2%, respectively. The proposed approach, combining both modalities, also provides increased accuracy compared to using only egovision data and only IMU data.

47 OTHER INSTRUMENTATION↗

Automated Labeling of Electron Microscopy Images Using Deep Learning

© 2018 IEEE. Searching for scientific data requires metadata providing a relevant context. Today, generating metadata is a time and labor intensive manual process that is often neglected, and important datasets are not accessible through search. We investigate the use of machine learning to generalize metadata from a subset of labeled data, thus increasing the availability of meaningful metadata for search. Specifically, we consider electron microscopy images collected at the National Center for Electron Microscopy (NCEM) at the Lawrence Berkeley National Laboratory (LBNL) and use deep learning to discern characteristics from a small subset of labeled images and transfer labels to the entire image corpus. Relatively small training set sizes and a minimum resolution of 512\times 512 pixels required by the application domain pose unique challenges. We overcome these challenges by using a simple yet powerful convolutional network architecture that limits the number of free parameters to lower the required amount of computational power and reduce the risk of overfitting. We achieve a classification accuracy of approximately 80% in discerning between images recorded in two operating modes of the electron microscope-transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM). We use transfer learning-i.e., re-using the pre-trained convolution layers from the TEM vs. STEM classification problem-to generalize labels and achieve an accuracy of approximately 70 % despite current experiments being limited to small training data sets. We present these predictions as suggestions to domain scientists through a web-based tool to accelerate the labeling process with the goal of further validating our approach and improving the accuracy of automatically created labels.

Weber, Gunther H↗

Fine-Grained Exploitation of Mixed Precision for Faster CNN Training

As deep convolutional neural networks (CNNs) have become increasingly popular and successful at an ever-widening number of machine learning tasks specialized hardware has become increasingly available for training and deploying them. NVIDIA's recent Volta architecture includes tensor cores which perform a fused operation reduced and mixed precision (16-bit multiply, 32-bit accumulate). Recent research indicates that, typically, very little is lost (in terms of training accuracy) when half precision is used in place of single precision, and performance gains can be made by doing arithmetic in reduced precision. In this work we demonstrate that making layer-by-layer choices as to the arithmetic/data precision can lead to further performance improvement. In our study of 25,200 CNNs we demonstrate an average speedup (over purely half precision) of 1.27x and speedups as high as 3.64x by appropriately combining single and half precision arithmetic and data types on a layer-by-layer basis.

Johnston, Travis↗

Reducing Communication in Graph Neural Network Training

Graph Neural Networks (GNNs) are powerful and flexible neural networks that use the naturally sparse connectivity information of the data. GNNs represent this connectivity as sparse matrices, which have lower arithmetic intensity and thus higher communication costs compared to dense matrices, making GNNs harder to scale to high concurrencies than convolutional or fully-connected neural networks. Here, we introduce a family of parallel algorithms for training GNNs and show that they can asymptotically reduce communication compared to previous parallel GNN training methods. We implement these algorithms, which are based on 1D, 1. 5D, 2D, and 3D sparse-dense matrix multiplication, using torch.distributed on GPU-equipped clusters. Our algorithms optimize communication across the full GNN training pipeline. We train GNNs on over a hundred GPUs on multiple datasets, including a protein network with over a billion edges.

97 MATHEMATICS AND COMPUTING↗

Multiscale Data-Driven Seismic Full-Waveform Inversion With Field Data Study

Seismic full-waveform inversion (FWI), which uses iterative methods to estimate high-resolution subsurface models from seismograms, is a powerful imaging technique in exploration geophysics. In recent years, the computational cost of FWI has grown exponentially due to the increasing size and resolution of seismic data. Moreover, it is a nonconvex problem and can encounter local minima due to the limited accuracy of the initial velocity models or the absence of low frequencies in the measurements. To overcome these computational issues, we develop a multiscale data-driven FWI method based on fully convolutional networks (FCNs). In preparing the training data, we first develop a real-time style transform method to create a large set of synthetic subsurface velocity models from natural images. We then develop two convolutional neural networks with encoder-decoder structures to reconstruct the low- and high-frequency components of the subsurface velocity models, separately. To validate the performance of our data-driven inversion method and the effectiveness of the synthesized training set, we compare it with conventional physics-based waveform inversion approaches using both synthetic and field data. Finally, these numerical results demonstrate that, once our model is fully trained, it can significantly reduce the computation time and yield more accurate subsurface velocity models in comparison with conventional FWI.

58 GEOSCIENCES↗

Iterative Reconstruction of Coded Source Neutron Radiographs

Use of a coded source facilitates high-resolution neutron imaging through magnifications but requires that the radiographic data be deconvolved. A comparison of direct deconvolution with two different iterative algorithms has been performed. One iterative algorithm is based on a maximum likelihood estimation (MLE)-like framework and the second is based on a geometric model of the neutron beam within a least squares formulation of the inverse imaging problem. Simulated data for both uniform and Gaussian shaped source distributions was used for testing to understand the impact of non-uniformities present in neutron beam distributions on the reconstructed images. Results indicate that the model based reconstruction method will match resolution and improve on contrast over convolution methods in the presence of non-uniform sources. Additionally, the model based iterative algorithm provides direct calculation of quantitative transmission values while the convolution based methods must be normalized base on known values.

Santos-Villalobos, Hector J↗