Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Matrix factorization”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Mapping the proteogenomic landscape enables prediction of drug response in acute myeloid leukemia

Acute myeloid leukemia is a poor prognosis cancer commonly stratified by genetic aberrations, but these mutations are often heterogeneous and don’t always predict therapeutic response. Here we combine transcriptomic, proteomic, and phosphoproteomic datasets with ex vivo drug sensitivity data to help understand the underlying pathophysiology of AML beyond mutations. We measured the proteome and phosphoproteome of 210 patients and combined them with genomics and transcriptomic measurements to identify four proteogenomic subtypes that complemented existing genetic subtypes. We then built a predictor to classify samples into subtypes based on 147 molecular features and mapped them to a ‘landscape’. Each region of this landscape corresponded to specific drug response patterns. We then built a drug response prediction model to identify drugs that target distinct subtypes. We can ultimately use these models to predict drug treatment response and prioritize treatments. Finally, we extended our models and mapped a series of cell lines representing various stages of quizartinib resistance into our subtype landscape, predicting and experimentally validating a switch in sensitivity to venetoclax to panobinostat, two drugs with very different mechanisms than quizartinib. Our results show how multi-omics data together with drug sensitivity data can inform therapy stratification and drug combinations in AML.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Python Codebase and Jupyter Notebooks - Applications of Machine Learning Techniques to Geothermal Play Fairway Analysis in the Great Basin Region, Nevada

Git archive containing Python modules and resources used to generate machine-learning models used in the "Applications of Machine Learning Techniques to Geothermal Play Fairway Analysis in the Great Basin Region, Nevada" project. This software is licensed as free to use, modify, and distribute with attribution. Full license details are included within the archive. See "documentation.zip" for setup instructions and file trees annotated with module descriptions.

Brown, Stephen↗
DOE OSTI2022

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↗
DOE OSTI2023

A General Method for Estimating Zonal Transmission Interface Limits from Nodal Network Data: Preprint

Capacity expansion models for the electric power system often employ zonal (rather than nodal) resolution, necessitating estimates of aggregate power transfer limits across the interfaces between model zones. Interface limits between planning areas are sometimes published, but they are not generalizable to arbitrary zone shapes. There is thus a need for a reproducible method for estimating interface transfer limits (ITLs) between user-defined zones directly from nodal transmission system data. Here, we present a simple method for estimating ITLs using a DC power flow approximation via the power transfer distribution factor (PTDF) matrix. Linear optimization is performed to identify the distribution of power flows that maximizes the total flow on interface-crossing lines, subject to individual line ratings, limits on bus injection/withdrawal, and the relationships among flows, injections, and withdrawals imposed by the PTDF matrix. We demonstrate the application of the method on a 134-zone ~65000-bus system, and we explore the influence of flow direction, contingency level, and zone size on the estimated ITLs. There is significant heterogeneity in the ratio of the ITL to the sum of interface-crossing line ratings, which highlights the importance of accounting for the physical constraints on power flows imposed by Kirchhoff's laws when estimating zonal ITLs.

17 WIND ENERGY↗
DOE OSTI2026

Fast and accurate calculation of EXAFS Debye-Waller factors in U⁢O2 using the dynamical matrix method

Theoretical modeling of bonding dynamics in metal oxides is required for predicting their thermal conductivity, catalytic activity, and mechanical properties. A primary challenge is the scarcity of experimental methods for validating theoretical predictions of these atomic-scale dynamics. This work presents a workflow that uses experimental extended x-ray absorption fine structure (EXAFS) data collected at high temperatures to validate an interatomic force field for uranium dioxide (UO2), an important model material. The validated force field is then used to drive computationally intensive molecular dynamics (MD) simulations and as input for the much faster dynamical matrix Debye-Waller (DMDW) method. The predicted values of the Debye-Waller factors from the DMDW calculations are in good agreement with those obtained from the MD simulations, with residual pair-specific differences attributable to quantum zero-point motion at low temperatures and lattice anharmonicity at high temperatures. We further show that theoretical EXAFS spectra constructed directly from DMDW-derived Debye-Waller factors reproduce the experimental data (at relatively low temperatures) with accuracy comparable to full MD-EXAFS, providing an additional validation of the choice of the potential. This study establishes a validated, rapid computational pathway for modeling bond dynamics, naturally incorporating quantum nuclear\\\\r\\\\nstatistics absent in classical simulations, which are essential for the mechanistic understanding of complex oxide materials.

58 GEOSCIENCES↗
DOE OSTI2022

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↗
DOE OSTI2023

Potential molecular targets for intervention in pelvic organ prolapse

Pelvic organ prolapse (POP) is a concerning gynecological benign illness in middle-aged and senior women. Its etiology is complex, the incidence rate is high, symptoms are clinically subjective, and its influence tends to be polarized. At present, for those who need medical treatment, whether surgical or non-surgical, complications cannot be ignored, and treatment effect needs to be optimized. However, there is a lack of accurate molecular biological interventions for the prevention, diagnosis, progression delay, and treatment of POP. Here, we reviewed the current state of understanding of the molecular mechanisms and factors associated with POP etiology. These factors include cyclins, matrix metal peptidases/tissue inhibitors of metalloproteinases, microRNAs, homeobox A11, transforming growth factor β1, insulin-like growth factor 1, fibulin 5, lysyl oxidase-like 1, oxidative stress, inflammatory response, estrogen, and other potential biomarkers associated with POP. In addition, relevant molecular targets that may be used to intervene in POP are summarized. The aim of this review was to provide more information to identify accurate potential biomarkers and/or molecular targets for the prevention, diagnosis, progression delay, and treatment of POP, with the goal of improving medical treatment for patients at-risk for POP or having POP. Continued research is needed to identify additional details of currently accepted molecular mechanisms and to identify additional mechanisms that contribute to POP.

Wu, Xia↗
DOE OSTI2020

Low synchronization Gram–Schmidt and generalized minimal residual algorithms

The Gram–Schmidt process uses orthogonal projection to construct the A = QR factorization of a matrix. When Q has linearly independent columns, the operator P = I - Q(QTQ)-1QT defines an orthogonal projection onto Q⊥. In finite precision, Q loses orthogonality as the factorization progresses. A family of approximate projections is derived with the form P = I - QTQT, with correction matrix T. When T = (QTQ)-1, and T is triangular, it is postulated that the best achievable orthogonality is $\mathcal{O}(ε)\mathcal{K}(A)$. We present new variants of modified (MGS) and classical Gram–Schmidt algorithms that require one global reduction step. An interesting form of the projector leads to a compact WY representation for MGS. In particular, the inverse compact WY MGS algorithm is equivalent to a lower triangular solve. Our main contribution is to introduce a backward normalization lag into the compact WY representation, resulting in a $\mathcal{O}(ε)\mathcal{K}[r_0, AV_m])$ stable Generalized Minimal Residual Method (GMRES) algorithm that requires only one global reduce per iteration. Finally, further improvements in performance are achieved by accelerating GMRES on GPUs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Fibrous anisotropy and mineral gradients within the radula stylus of chiton: Controlled stiffness and damage tolerance in a flexible biological composite

Over hundreds of millions of years, organisms have evolved architected structures via precise control over hierarchically assembled components, including the integration of dissimilar materials. One such example is found in the radula system of chitons, intertidal mollusks that feed on algae growing on the rock. Their radula consists of multiple rows of ultrahard teeth, each integrated with a foldable belt-like substrate via a stiff, yet flexible stylus, which is essential for efficient rasping during the feeding process. Here, we investigate the nano and micro-scale components and architectures as well as regional mechanical properties of the stylus, and their subsequent role during the rasping of Cryptochiton stelleri. Three important factors were determined to contribute to the regio-specific stiffness of the stylus: the presence of mineral components, highly oriented chitinous fibers, and a chemically cross-linked protein matrix. All these factors are varied throughout the stylus. There is a high mineral content on the trailing edge close to the tooth and a cross-linked matrix on the leading edge, both with orientational specific oriented chitin fibers that provide force transduction to the tooth. Conversely, there is a significant lack of mineral or cross-linked matrix in the proximal end as well as a low degree of fiber orientation, resulting in a flexible region that can accommodate torsion and flexure during rasping. Understanding the graded composite structure of the stylus and applying this unique design to various engineering fields such as soft robotics, biotechnology, and the medical industry, can inspire the production of high-performance materials.

Lee, Jung-Eun↗
DOE OSTI2020

Finite Element Modeling of the Fiber-Matrix Interface in Polymer Composites

Polymer composites are used in numerous industries due to their high specific strength and high specific stiffness. Composites have markedly different properties than both the reinforcement and the matrix. Of the several factors that govern the final properties of the composite, the interface is an important factor that influences the stress transfer between the fiber and matrix. The present study is an effort to characterize and model the fiber-matrix interface in polymer matrix composites. Finite element models were developed to study the interfacial behavior during pull-out of a single fiber in continuous fiber-reinforced polymer composites. A three-dimensional (3D) unit-cell cohesive damage model (CDM) for the fiber/matrix interface debonding was employed to investigate the effect of interface/sizing coverage on the fiber. Furthermore, a two-dimensional (2D) axisymmetric model was used to (a) analyze the sensitivity of interface stiffness, interface strength, friction coefficient, and fiber length via a parametric study; and (b) study the shear stress distribution across the fiber-interface-matrix zone. It was determined that the force required to debond a single fiber from the matrix is three times higher if there is adequate distribution of the sizing on the fiber. The parametric study indicated that cohesive strength was the most influential factor in debonding. Moreover, the stress distribution model showed the debonding mechanism of the interface. It was observed that the interface debonded first from the matrix and remained in contact with the fiber even when the fiber was completely pulled out.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Connecting Matrix Elements to Multi-Hadron Form-Factors

We discuss developments in calculating multi-hadron form-factors and transition processes via lattice QCD. Our primary tools are finite-volume scaling relations, which map spectra and matrix elements to the corresponding multi-hadron infinite-volume amplitudes. We focus on two hadron processes probed by an external current, and provide various checks on the finite-volume formalism in the limiting cases of perturbative interactions and systems forming a bound state. By studying model-independent properties of the infinite-volume amplitudes, we are able to rigorously define form-factors of resonances.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

GPU Accelerated Sparse Cholesky Factorization

The solution of sparse symmetric positive definite linear systems is an important computational kernel in large-scale scientific and engineering modeling and simulation. We will solve the linear systems using a direct method, in which a Cholesky factorization of the coefficient matrix is performed using a right-looking approach and the resulting triangular factors are used to compute the solution. Sparse Cholesky factorization is compute intensive. In this work we investigate techniques for reducing the factorization time in sparse Cholesky factorization by offloading some of the dense matrix operations on a GPU. We will describe the techniques we have considered. We achieved up to 4x speedup compared to the CPU-only version.

Karsavuran, M Ozan↗
DOE OSTI2026

Quantum block encoding for one-pair semiseparable matrices

Quantum block encoding (QBE) is a crucial step in the development of most quantum algorithms, as it provides an embedding of a given matrix into a suitable larger unitary matrix. Historically, the development of efficient techniques for QBE has mostly focused on sparse matrices; less effort has been devoted to data-sparse (e.g., rank-structured) matrices. In this work we examine a particular case of rank structure, namely, one-pair semiseparable matrices. We present a new block encoding approach that relies on a suitable factorization of the given matrix as the product of triangular and diagonal factors. To encode the matrix, the algorithm needs $2\log(N)+7$ ancillary qubits. Assuming that the data input oracles can be implemented with polylogarithmic depth, or that a QRAM input model is available, our proposed method requires $\mathcal{O}({\rm polylog} (N))$ time and has an error of $\mathcal{O}(N^2)$, where $N$ is the matrix size.

Antonioli, Giacomo [Pisa U.; CERN] (ORCID:00090000↗