Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Network inference”

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 397 records · Page 22
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Wilson loops with neural networks

Wilson loops are essential objects in QCD and have been pivotal in scale setting and demonstrating confinement. Various generalizations are crucial for computations needed in effective field theories. In lattice gauge theory, Wilson loop calculations face challenges, including excited-state contamination at short times and the signal-to-noise ratio issue at longer times. To address these problems, we develop a new method by using neural networks to parametrize interpolators for the static quark-antiquark pair. We construct gauge-equivariant layers for the network and train it to find the ground state of the system. The trained network itself is then treated as our new observable for the inference. Our results demonstrate a significant improvement in the signal compared to traditional Wilson loops, performing as well as Coulomb-gauge Wilson-line correlators while maintaining gauge invariance. Additionally, we present an example where the optimized ground state is used to measure the static force directly, as well as another example combining this method with the multilevel algorithm. Finally, we extend the formalism to find excited-state interpolators for static quark-antiquark systems. To our knowledge, this work is the first study of neural networks with a physically motivated loss function for Wilson loops.

Bellscheidt, Verena [Massachusetts Inst. of Techno↗
DOE OSTI2023

Watersheds and stream networks viewed longitudinally: Example insights from novel spatial portrayals of watershed characteristics

Abstract Longitudinal depictions of watershed structure and characteristics, including topography, stream networks, wetlands, ground water levels, and land use, can provide watershed knowledge and understanding unavailable from standard plan view maps. Three case studies provide examples of knowledge gained by applying longitudinal views of stream networks, watershed hydrologic behavior, and land use distributions. Longitudinal views of mountain stream networks show extreme variability in the slope‐area relationships of low Strahler order streams, large discontinuities in drainage area (large parts of drainage area space are absent in networks), and large variations in network curvature. Longitudinal views of a groundwater‐dominated headwater watershed increase the inference available from limited groundwater observations and clearly reveal how groundwater connections affect the permanence of surface water features and the distribution of vadose zone storage in the landscape. Plotting land uses longitudinally illuminates and allows a quantitative analysis of how land uses are distributed relative to topographic position. Viewing watersheds and stream networks longitudinally can provide new insights into watershed forms and processes and motivate new questions and research.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Feasibility of Vascular Parameter Estimation for Assessing Hypertensive Pregnancy Disorders

Abstract Hypertensive pregnancy disorders (HPDs), such as pre-eclampsia, are leading sources of both maternal and fetal morbidity in pregnancy. Noninvasive imaging, such as ultrasound (US) and magnetic resonance imaging (MRI), is an important tool for predicting and monitoring these high risk pregnancies. While imaging can measure hemodynamic parameters, such as uterine artery pulsatility and resistivity indices (PI and RI), the interpretation of such metrics for disease assessment relies on ad hoc standards, which provide limited insight to the physical mechanisms underlying the emergence of hypertensive pregnancy disorders. To provide meaningful interpretation of measured hemodynamic data in patients, advances in computational fluid dynamics can be brought to bear. In this work, we develop a patient-specific computational framework that combines Bayesian inference with a reduced-order fluid dynamics model to infer parameters, such as vascular resistance, compliance, and vessel cross-sectional area, known to be related to the development of hypertension. The proposed framework enables the prediction of hemodynamic quantities of interest, such as pressure and velocity, directly from sparse and noisy MRI measurements. We illustrate the effectiveness of this approach in two systemic arterial network geometries: an aorta with branching carotid artery and a maternal pelvic arterial network. For both cases, the model can reconstruct the provided measurements and infer parameters of interest. In the case of the maternal pelvic arteries, the model can make a distinction between the pregnancies destined to develop hypertension and those that remain normotensive, expressed through the value range of the predicted absolute pressure.

Biophysics↗
DOE OSTI2020

ContourNet: Salient Local Contour Identification for Blob Detection in Fusion Plasma Simulation Data

We present ContourNet, a deep learning approach to identify salient local isocontours as blobs in large-scale 5D gyrokinetic tokamak simulation data. Blobs-regions of high turbulence that run along the edge wall down toward the diverter and can damage the tokamak-are non-well-defined features but have been empirically localized by isocontours in 2D normalized fluctuating density fields. The key of our study is to train ContourNet to follow the empirical rules to detect blobs over the time-varying simulation data. The architecture of ContourNet is a convolutional neural segmentation network: the inputs are the density field and a rasterized isocontour; the output is a set of isocontour encircling blobs. At the training stage, we feed the network with manually identified isocontours and propagated labels. At the inference stage, we extract isocontours from the segmented blob regions. Results show that our approach can achieve both high accuracy and performance, which enables scientists to understand the blob dynamics influencing the confinement of the plasma.

Blob detection↗
DOE OSTI2025

A small core in Vesta inferred from Dawn’s observations

Vesta’s large-scale interior structure had previously been constrained primarily using the gravity and shape data from the Dawn mission. However, these data alone still allow a wide range of possibilities for the differentiation state of the body. The moment of inertia is arguably the most diagnostic parameter related to the radial density distribution of a planetary body, making it crucial for assessing the body’s state of internal differentiation. Determining the moment of inertia requires additional measurements of the amplitudes of small rotational motions, such as precession and nutation. Here we report an updated estimate of the moment of inertia of Vesta inferred from Dawn’s Doppler tracking via the Deep Space Network and onboard imaging data. The recovered value for Vesta’s normalized polar moment of inertia is $\overline{C}$/MR 2 = 0.4208 ± 0.0047 (where M is the mass of Vesta and R is the reference radius), which is only 6.6% lower than the homogeneous value of 0.4505. This value, combined with the gravity field and global shape, suggests that Vesta’s interior has limited density stratification beneath its howardite–eucrite–diogenite-dominated crust. We propose two possible origin scenarios that are consistent with the observed constraints. In the first scenario, Vesta’s interior did not undergo full differentiation due to late accretion. In the second scenario, Vesta originated as an impact remnant of a larger differentiated body re-accreted with non-chondritic bulk composition produced from a catastrophic impact. Vesta did not experience complete differentiation in either scenario, suggesting that its current state reflects a complex interplay between its accretion timing, thermal evolution, redistribution of 26 Al bearing melt and/or impact processes.

CNEOS 2014-01-08 bolide↗
DOE OSTI2020

The changing mouse embryo transcriptome at whole tissue and single-cell resolution

During mammalian embryogenesis, differential gene expression gradually builds the identity and complexity of each tissue and organ system 1 . Here we systematically quantified mouse polyA-RNA from day 10.5 of embryonic development to birth, sampling 17 tissues and organs. The resulting developmental transcriptome is globally structured by dynamic cytodifferentiation, body-axis and cell-proliferation gene sets that were further characterized by the transcription factor motif codes of their promoters. We decomposed the tissue-level transcriptome using single-cell RNA-seq (sequencing of RNA reverse transcribed into cDNA) and found that neurogenesis and haematopoiesis dominate at both the gene and cellular levels, jointly accounting for one-third of differential gene expression and more than 40% of identified cell types. By integrating promoter sequence motifs with companion ENCODE epigenomic profiles, we identified a prominent promoter de-repression mechanism in neuronal expression clusters that was attributable to known and novel repressors. Focusing on the developing limb, single-cell RNA data identified 25 candidate cell types that included progenitor and differentiating states with computationally inferred lineage relationships. We extracted cell-type transcription factor networks and complementary sets of candidate enhancer elements by using single-cell RNA-seq to decompose integrative cis -element (IDEAS) models that were derived from whole-tissue epigenome chromatin data. These ENCODE reference data, computed network components and IDEAS chromatin segmentations are companion resources to the matching epigenomic developmental matrix, and are available for researchers to further mine and integrate.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Ultrafast jet classification at the HL-LHC

Abstract Three machine learning models are used to perform jet origin classification. These models are optimized for deployment on a field-programmable gate array device. In this context, we demonstrate how latency and resource consumption scale with the input size and choice of algorithm. Moreover, the models proposed here are designed to work on the type of data and under the foreseen conditions at the CERN large hadron collider during its high-luminosity phase. Through quantization-aware training and efficient synthetization for a specific field programmable gate array, we show that O ( 100 ) ns inference of complex architectures such as Deep Sets and Interaction Networks is feasible at a relatively low computational resource cost.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Spy in the GPU-box: Covert and Side Channel Attacks on Multi-GPU System

The deep learning revolution has been enabled in large part by GPUs, and more recently accelerators, which make it possible to carry out computationally demanding training and inference in acceptable times. As the size of machine learning networks and workloads continues to increase, multi-GPU machines have emerged as an important platform offered on High Performance Computing and cloud data centers. Since these machines are shared among multiple users, it becomes increasingly important to protect applications against potential attacks. In this paper, we explore the vulnerability of Nvidia's DGX multi-GPU machines to covert and side channel attacks. These machines consist of a number of discrete GPUs that are interconnected through a combination of custom interconnect (NVLink) and PCIe connections. We reverse engineer the interconnected cache hierarchy and show that it is possible for an attacker on one GPU to cause contention on the L2 cache of another GPU. We use this observation to first develop a covert channel attack across two GPUs, achieving the best bandwidth of around 4 MB/s. We also develop a prime and probe attack on a remote GPU allowing an attacker to recover the cache access pattern of another workload. This access pattern can be used in any number of side channel attacks: we demonstrate a proof of concept attack that fingerprints the application running on the remote GPU, with high accuracy. We also develop a proof of concept attack to extract hyperparameters of a machine learning workload. Our work establishes for the first time the vulnerability of these machines to microarchitectural attacks and can guide future research to improve their security.

Dutta, Sankha↗
DOE OSTI2020

DiHydrogen

DiHydrogen is the second version of the Hydrogen fork of the well-known distributed linear algebra library, Elemental. DiHydrogen is a GPU-accelerated distributed multilinear algebra interface with a particular emphasis on the needs of the scalable distributed deep learning training and inference. DiHydrogen is part of the Livermore Big Artificial Neural Network (LBANN) software stack.

Maruyama, Naoya↗
DOE OSTI2026

Reconstructing the Stripping History of the Sagittarius Stream with Neural Networks

The Sagittarius (Sgr) Stream is produced by the ongoing disruption of the Sgr dwarf spheroidal (dSph) galaxy and is thought to contain multiple wraps that were stripped during different pericentric passages. In this study, we introduce a neural-network–based method trained on N-body simulations to infer the stripping time of Sgr Stream stars directly from their phase-space coordinates. We combine spectroscopic data from SEGUE, APOGEE DR17, and LAMOST DR7 low-resolution spectroscopic (LRS) survey with Gaia EDR3 astrometry and distance estimates from the latest StarHorse catalog to identify high-quality Sgr Stream members. Applying our method to these stars, we measure a clear metallicity gradient with stripping time, well described by a linear relation with slope ∼0.3 dex Gyr −1 . We further predict the stripping times of globular clusters previously suggested to originate from the Sgr dSph. M 54, Terzan 7, Terzan 8, and Arp 2 exhibit stripping times consistent with being currently bound to the Sgr remnant. Pal 12, Whiting 1, and NGC 2419 are inferred to have been stripped 0.9 ± 0.1, 1.1 ± 0.2, and 2.1 ± 0.2 Gyr ago, respectively. For NGC 4147 and NGC 5634, whose membership in the Sgr system remains uncertain, our analysis suggests stripping times of 1.1 ± 0.4 and 1.1 ± 0.1 Gyr, respectively, if they are ultimately confirmed as genuine Sgr members. These results demonstrate that data-driven models of dynamical stripping histories offer a promising approach for reconstructing the formation and chemical evolution of the Sgr Stream.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Aided Active Learning (AAL) for Enhanced Critical Heat Flux Prediction

Accurate prediction of critical heat flux (CHF) is crucial for the safe and efficient operation of nuclear reactors. Traditional CHF modeling methods often require extensive experimental data, which are hard to obtain. This study introduces the Aided Active Learning (AAL) framework, which strategically minimizes data requirements without sacrificing model accuracy. Unlike conventional Active Learning (AL), AAL introduces an additional step of randomly selecting a subset from the sample pool before applying the query strategy. To evaluate the performance of AAL, two query strategies—uncertainty-based sampling and error-reduction sampling—were evaluated across the following models: random forest (RF), feedforward neural network (FNN), and variational feedforward neural network (vFNN). The proposed framework demonstrated that AAL effectively reduces the number of training samples needed to achieve comparable predictive accuracy. For the RF model, AL required only 710 samples to achieve an R2 score of 0.98, as compared to the 4,785 samples needed by random sampling. Similarly, the FNN model achieved the same R2 score with just 355 samples when using AL, a significant improvement over the 825 samples required by random sampling. In case of uncertainty-based sampling strategy, vFNN attained an R2 of 0.98 with 3,420 samples, reducing the sample requirement by 47% relative to the 6,440 samples needed for random sampling. Its performance suggests that larger training data are required to fully leverage its uncertainty quantification capabilities.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Kilonova light-curve interpolation with neural networks

Kilonovae are the electromagnetic transients created by the radioactive decay of freshly synthesized elements in the environment surrounding a neutron star merger. To study the fundamental physics in these complex environments, kilonova modeling requires, in part, the use of radiative transfer simulations. The microphysics involved in these simulations results in high computational cost, prompting the use of emulators for parameter inference applications. Utilizing a training set of 22 248 high-fidelity simulations (composed of 412 unique ejecta parameter combinations evaluated at 54 viewing angles), we use a neural network to efficiently train on existing radiative transfer simulations and predict light curves for new parameters in a fast and computationally efficient manner. Our neural network can generate millions of new light curves in under a minute. We discuss our emulator's degree of off-sample reliability and parameter inference of the AT2017gfo observational data. Finally, we discuss tension introduced by multiband inference in the parameter inference results, particularly with regard to the neural network's recovery of viewing angle. Published by the American Physical Society 2024

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

A Quantum-Based Approach to Predict Primary Radiation Damage in Polymeric Networks

Initial atomistic-level radiation damage in chemically reactive materials is thought to induce reaction cascades that can result in undesirable degradation of macroscale properties. Ensembles of quantum-based molecular dynamics (QMD) simulations can accurately predict these cascades, but extracting chemical insights from the many underlying trajectories is a labor-intensive process that can require substantial a priori intuition. We develop here a general and automated graph-based approach to extract all chemically distinct structures sampled in QMD simulations and apply our approach to predict primary radiation damage of polydimethylsiloxane (PDMS), the main constituent of silicones. A postprocessing protocol is developed to identify underlying polymer backbone structures as connected components in QMD trajectories. Furthermore, these backbones form a repository of radiation-damaged structures. A scheme for extracting and updating a library of isomorphically distinct structures is proposed to identify the spanning set and aid chemical interpretation of the repository. The analyses are applied to ensembles of cascade QMD simulations in which the four element types in PDMS are selectively excited in primary knock-on atom events. Our approach reveals a much higher degree of combinatorial complexity in this system than was inferred through radiolysis experiments. Probabilities are extracted for radiation-induced network changes including formation of branch points, carbon linkages, cycles, bond scissions, and carbon uptake into the Si–O siloxane backbone network. The general analysis framework presented here is readily extendable to modeling chemical degradation of other polymers and molecular materials and provides a basis for future quantum-informed multiscale modeling of radiation damage.

36 MATERIALS SCIENCE↗
DOE OSTI2026

E-PINNs: Epistemic Physics-Informed Neural Networks

Physics-informed neural networks (PINNs) have demonstrated promise as a framework for solving forward and inverse problems involving partial differential equations. Despite recent progress in the field, it remains challenging to quantify uncertainty in these networks. While techniques such as Bayesian PINNs (B-PINNs) provide a principled approach to capturing epistemic uncertainty through Bayesian inference, they can be computationally expensive for large-scale applications. In this work, we propose Epistemic Physics-Informed Neural Networks (E-PINNs), a framework that uses a small network, the epinet, to efficiently quantify epistemic uncertainty in PINNs. The proposed approach works as an add-on to existing, pre-trained PINNs with a small computational overhead. We demonstrate the applicability of the proposed framework in various test cases and compare the results with B-PINNs using Hamiltonian Monte Carlo (HMC) posterior estimation and dropout-equipped PINNs (Dropout-PINNs). In our experiments, E-PINNs achieve calibrated coverage with competitive sharpness at substantially lower cost. We demonstrate that when B-PINNs produce narrower bands, they under-cover in our tests. E-PINNs also show better calibration than Dropout-PINNs in these examples, indicating a favorable accuracy-efficiency trade-off.

AI for Science↗
DOE OSTI2023

Tencoder: tensor-product encoder-decoder architecture for predicting solutions of PDEs with variable boundary data

It is widely hoped that artificial intelligence will boost data-driven surrogate models in science and engineering. However, fundamental spatial aspects of AI surrogate models remain under-studied. We investigate the ability of neural-network surrogate models to predict solutions to PDEs under variable boundary values. We do not wish to retrain the model when the boundary values change but to make them inputs to the model and infer the solution of the PDE under those boundary conditions. Such a capability is essential to making AI-based surrogate models practically useful. While simple feedforward networks are used for one-dimensional (1D) Poisson equation, an encoder-decoder architecture with a tensor-product layer is developed for the two-dimensional Poisson equation posed on a rectangular domain. We show that it is indeed possible to infer solutions to PDEs from variable boundary data using neural networks in this relatively simple setting, and point to future directions.

Kashi, Aditya↗
NASA NTRS1972

Large-scale solar magnetic fields and H-alpha patterns

Coronal and interplanetary magnetic fields computed from measurements of large-scale photospheric magnetic fields suffer from interruptions in day-to-day observations and the limitation of using only measurements made near the solar central meridian. Procedures were devised for inferring the lines of polarity reversal from H-alpha solar patrol photographs that map the same large-scale features found on Mt. Wilson magnetograms. These features may be monitored without interruption by combining observations from the global network of observatories associated with NOAA's Space Environment Services Center. The patterns of inferred magnetic fields may be followed accurately as far as 60 deg from central meridian. Such patterns will be used to improve predictions of coronal features during the next solar eclipse.

Mcintosh, P. S.↗
DOE OSTI2021

Quantifying microbial control of soil organic matter dynamics at macrosystem scales

Soil organic matter (SOM) stocks, decomposition and persistence are largely the product of controls that act locally. Yet the controls are shaped and interact at multiple spatiotemporal scales, from which macrosystem patterns in SOM emerge. Theory on SOM turnover recognizes the resulting spatial and temporal conditionality in the effect sizes of controls that play out across macrosystems, and couples them through evolutionary and community assembly processes. For example, climate history shapes plant functional traits, which in turn interact with contemporary climate to influence SOM dynamics. Selection and assembly also shape the functional traits of soil decomposer communities, but it is less clear how in turn these traits influence temporal macrosystem patterns in SOM turnover. Here, we review evidence that establishes the expectation that selection and assembly should generate decomposer communities across macrosystems that have distinct functional effects on SOM dynamics. Representation of this knowledge in soil biogeochemical models affects the magnitude and direction of projected SOM responses under global change. Yet there is high uncertainty and low confidence in these projections. To address these issues, we make the case that a coordinated set of empirical practices are required which necessitate (1) greater use of statistical approaches in biogeochemistry that are suited to causative inference; (2) long-term, macrosystem-scale, observational and experimental networks to reveal conditionality in effect sizes, and embedded correlation, in controls on SOM turnover; and (3) use of multiple measurement grains to capture local- and macroscale variation in controls and outcomes, to avoid obscuring causative understanding through data aggregation. Here, when employed together, along with process-based models to synthesize knowledge and guide further empirical work, we believe these practices will rapidly advance understanding of microbial controls on SOM and improve carbon cycle projections that guide policies on climate adaptation and mitigation.

59 BASIC BIOLOGICAL SCIENCES↗