Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed 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 595 records · Page 33
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Estimating Cosmological Constraints from Galaxy Cluster Abundance using Simulation-Based Inference

Inferring the values and uncertainties of cosmological parameters in a cosmology model is of paramount importance for modern cosmic observations. In this paper, we use the simulation-based inference (SBI) approach to estimate cosmological constraints from a simplified galaxy cluster observation analysis. Using data generated from the Quijote simulation suite and analytical models, we train a machine learning algorithm to learn the probability function between cosmological parameters and the possible galaxy cluster observables. The posterior distribution of the cosmological parameters at a given observation is then obtained by sampling the predictions from the trained algorithm. Our results show that the SBI method can successfully recover the truth values of the cosmological parameters within the 2σ limit for this simplified galaxy cluster analysis, and acquires similar posterior constraints obtained with a likelihood-based Markov Chain Monte Carlo method, the current state-of the-art method used in similar cosmological studies.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

A segmented approach to modeling building height: Delineating high-rise and low-rise buildings for enhanced height estimation

Understanding building height is imperative to the overall study of energy efficiency, population distribution, urban morphologies, emergency response, among others. Currently, existing approaches for modeling building height at scale are hindered by two pervasive issues. First, there is no consistent approach to quantify what a high-rise building is at a macro scale, leaving researchers unable to accurately compare results across geographies and domains. Second, high-rise buildings represent a small fraction of the built environment, implying data imbalance challenges that negatively affect current approaches. This is a problem of practical relevance since information on high-rise buildings is important for studies on urban heat islands, population dynamics, and pollution dispersion. Here, we introduce a novel approach to map building height which first identifies two distinct distributions within the built environment, with one being composed of low-rise buildings and one composed of high-rise buildings. We then develop an ensemble scheme where discrete specialist models are trained for each subset of low-rise buildings and high-rise buildings to infer building height from morphology features. For experiments mapping heights of 4.85 million buildings in Japan, we show an increase of 34 % in accuracy within 3m error when compared to the current state-of-the-art when modeling high-rise buildings, which based on KNN experimentation we define as any building > 12m . Our findings show that such an ensemble framework outperforms the current state-of-the-art approaches, which is especially relevant in relation to inferring height for high-rise buildings, a prominent issue of existing approaches for mapping the built environment.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Exploring the Use of Non‐Invasive Drone‐Based Ground‐Penetrating Radar (GPR) to Characterize Biogenic Gas Dynamics in Subtropical Peat Soils

Peat soils are a critical component of the global carbon cycle as natural producers of biogenic greenhouse gases (e.g., methane and carbon dioxide) that accumulate within the soil and are released to the atmosphere. Previous studies have showed the ability of ground-based minimally-invasive geophysical methods such as ground-penetrating radar (GPR) to characterize carbon dynamics in peat soils. However, ground-based GPR is limited by scale of measurement and soil disturbance potentially altering gas releases during deployment. Here, we explore the potential of drone-based GPR for identification of hot spots and hot moments of gas accumulation and release in subtropical soils. Here, we collected drone-based GPR data sets across two grids (∼17,500 m 2 ) in the Everglades during January (dry season), September, and November (wet season) of 2023 to characterize peat thickness and seasonal variability of gas content. Results show that drone-based GPR is effective and efficient for: (a) capturing the temporal variation of in situ biogenic gas content in peat soils with changes between 1% and 25 % volumetric gas content over repeatable grids; (b) inferring a total peat thickness between 0.8 and 1.2 m; and (c) estimating flux releases of 63 and 135 mg CH 4 m −2 day −1 for specific locations and periods that are strikingly consistent with our coincident gas trap measurements. This work also indicates that (a) spatial distribution of gas content in the Everglades is strongly controlled by landscape morphology such as ridges and sloughs and (b) the temporal variation of gas content is seasonal with increased gas production during the wet season.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Forecasting constraints on the mean free path of ionizing photons at z ≥ 5.4 from the Lyman-α forest flux autocorrelation function

ABSTRACT Fluctuations in Lyman-α (Ly α) forest transmission towards high-z quasars are partially sourced from spatial fluctuations in the ultraviolet background, the level of which are set by the mean free path of ionizing photons (λmfp). The autocorrelation function of Ly α forest flux characterizes the strength and scale of transmission fluctuations and, as we show, is thus sensitive to λmfp. Recent measurements at z ∼ 6 suggest a rapid evolution of λmfp at z > 5.0 which would leave a signature in the evolution of the autocorrelation function. For this forecast, we model mock Ly α forest data with properties similar to the XQR-30 extended data set at 5.4 ≤ z ≤ 6.0. At each z, we investigate 100 mock data sets and an ideal case where mock data matches model values of the autocorrelation function. For ideal data with λmfp = 9.0 cMpc at z = 6.0, we recover $\lambda _{\text{mfp}}=12^{+6}_{-3}$ cMpc. This precision is comparable to direct measurements of λmfp from the stacking of quasar spectra beyond the Lyman limit. Hypothetical high-resolution data leads to a $\sim 40~{{\ \rm per\ cent}}$ reduction in the error bars over all z. The distribution of mock values of the autocorrelation function in this work is highly non-Gaussian for high-z, which should caution work with other statistics of the high-z Ly α forest against making this assumption. We use a rigorous statistical method to pass an inference test, however future work on non-Gaussian methods will enable higher precision measurements.

Astronomy & Astrophysics↗
DOE OSTI2021

Combining Electromagnetic and Gravitational-Wave Constraints on Neutron-Star Masses and Radii

In this work, we perform a joint Bayesian inference of neutron-star mass and radius constraints based on GW170817, observations of quiescent low-mass x-ray binaries (QLMXBs), photospheric radius expansion x-ray bursting sources, and x-ray timing observations of J0030+0451. With this dataset, the form of the prior distribution still has an impact on the posterior mass-radius curves and equation of state (EOS), but this impact is smaller than recently obtained when considering QLMXBs alone. We analyze the consistency of the electromagnetic data by including an “intrinsic scattering” contribution to the uncertainties, and find only a slight broadening of the posteriors. This suggests that the gravitational-wave and electromagnetic observations of neutron-star structure are providing a consistent picture of the neutron-star mass-radius curve and the EOS.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

The DESI-Lensing Mock Challenge: large-scale cosmological analysis of 3x2-pt statistics

The current generation of large galaxy surveys will test the cosmological model by combining multiple types of observational probes. Realising the statistical promise of these new datasets requires rigorous attention to all aspects of analysis including cosmological measurements, modelling, covariance and parameter likelihood. In this paper we present the results of an end-to-end simulation study designed to test the analysis pipeline for the combination of the Dark Energy Spectroscopic Instrument (DESI) Year 1 galaxy redshift dataset and separate weak gravitational lensing information from the Kilo-Degree Survey, Dark Energy Survey and Hyper-Suprime-Cam Survey. Our analysis employs the 3x2-pt correlation functions including cosmic shear and galaxy-galaxy lensing, together with the projected correlation function of the spectroscopic DESI lenses. We build realistic simulations of these datasets including galaxy halo occupation distributions, photometric redshift errors, weights, multiplicative shear calibration biases and magnification. We calculate the analytical covariance of these correlation functions including the Gaussian, noise and super-sample contributions, and show that our covariance determination agrees with estimates based on the ensemble of simulations. We use a Bayesian inference platform to demonstrate that we can recover the fiducial cosmological parameters of the simulation within the statistical error margin of the experiment, investigating the sensitivity to scale cuts. This study is the first in a sequence of papers in which we present and validate the large-scale 3x2-pt cosmological analysis of DESI-Y1.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Effect of Extinction on Quasar Luminosity Distances Determined from UV and X-Ray Flux Measurements

Abstract In Khadka et al., a sample of X-ray-detected reverberation-mapped quasars was presented and applied for the comparison of cosmological constraints inferred using two well-established relations in active galactic nuclei—the X-ray/UV luminosity ( L X – L UV ) relation and the broad-line region radius–luminosity ( R – L ) relation. L X – L UV and R – L luminosity distances to the same quasars exhibit a distribution of their differences that is generally asymmetric and positively shifted for the six cosmological models we consider. We demonstrate that this behavior can be interpreted qualitatively as arising as a result of the dust extinction of UV/X-ray quasar emission. We show that the extinction always contributes to the nonzero difference between L X – L UV -based and R – L -based luminosity distances and we derive a linear relationship between the X-ray/UV color index E X −UV and the luminosity-distance difference, which also depends on the value of the L X – L UV relation slope. Taking into account the median and the peak values of the luminosity-distance difference distributions, the average X-ray/UV color index falls in the range of E ¯ X − UV = 0.03 – 0.28 mag for the current sample of 58 sources. This amount of extinction is typical for the majority of quasars and can be attributed to the circumnuclear and interstellar media of host galaxies. After applying the standard hard X-ray and far-UV extinction cuts, heavily extincted sources are removed but overall the shift toward positive values persists. The effect of extinction on luminosity distances is more pronounced for the L X – L UV relation since the extinction of UV and X-ray emissions both contribute.

Astronomy & Astrophysics↗
DOE OSTI2024

Integrated analysis of X-ray diffraction patterns and pair distribution functions for machine-learned phase identification

Abstract To bolster the accuracy of existing methods for automated phase identification from X-ray diffraction (XRD) patterns, we introduce a machine learning approach that uses a dual representation whereby XRD patterns are augmented with simulated pair distribution functions (PDFs). A convolutional neural network is trained directly on XRD patterns calculated using physics-informed data augmentation, which accounts for experimental artifacts such as lattice strain and crystallographic texture. A second network is trained on PDFs generated via Fourier transform of the augmented XRD patterns. At inference, these networks classify unknown samples by aggregating their predictions in a confidence-weighted sum. We show that such an integrated approach to phase identification provides enhanced accuracy by leveraging the benefits of each model’s input representation. Whereas networks trained on XRD patterns provide a reciprocal space representation and can effectively distinguish large diffraction peaks in multi-phase samples, networks trained on PDFs provide a real space representation and perform better when peaks with low intensity become important. These findings underscore the importance of using diverse input representations for machine learning models in materials science and point to new avenues for automating multi-modal characterization.

36 MATERIALS SCIENCE↗
DOE OSTI2022

The primordial matter power spectrum on sub-galactic scales

The primordial matter power spectrum quantifies fluctuations in the distribution of dark matter immediately following inflation. Over cosmic time, overdense regions of the primordial density field grow and collapse into dark matter haloes, whose abundance and density profiles retain memory of the initial conditions. By analysing the image magnifications in 11 strongly lensed and quadruply imaged quasars, we infer the abundance and concentrations of low-mass haloes, and cast the measurement in terms of the amplitude of the primordial matter power spectrum. We anchor the power spectrum on large scales, isolating the effect of small-scale deviations from the Lambda cold dark matter (ΛCDM) prediction. Assuming an analytic model for the power spectrum and accounting for several sources of potential systematic uncertainty, including three different models for the halo mass function, we obtain correlated inferences of $\log _{10}\left(P / P_{\Lambda \rm {CDM}}\right)$, the power spectrum amplitude relative to the predictions of the concordance cosmological model, of $0.0_{-0.4}^{+0.5}$, $0.1_{-0.6}^{+0.7}$, and $0.2_{-0.9}^{+1.0}$ at k = 10, 25, and 50 $\rm {Mpc^{-1}}$ at $68 {{\ \rm per\ cent}}$ confidence, consistent with CDM and single-field slow-roll inflation.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Learning Global Proliferation Expertise Evolution Using AI-Driven Analytics and Public Information

Detecting and anticipating global proliferation expertise and capability evolution from unstructured, noisy, and incomplete public data streams is a highly desired, but extremely challenging task. Here, in this article, we present our pioneering data-driven approach to support the non-proliferation mission to detect and explain the evolution of proliferation expertise and capability development globally from terabytes of publicly available information (PAI), focusing on our knowledge extraction pipeline and descriptive analytics. We first discuss how we fuse nine open-source data streams, including multilingual data, to convert 4 TB of unstructured data to structured knowledge and encode dynamically evolving proliferation expertise representations—content and context graphs. For this, we rely on natural language processing (NLP) and deep learning (DL) models to perform information extraction, topic modeling, and distributed text representation (aka embedding) learning. We then present interactive, usable, and explainable descriptive analytics to refine domain knowledge and present it in a human-understandable form. Finally, we introduce future work avenues that will leverage our dynamic knowledge representations and descriptive analytics to enable predictive and prescriptive inferences to achieve real-time domain understanding and contextual reasoning about global proliferation expertise and capability evolution.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2025

Intrepid MCMC: Metropolis-Hastings with exploration

In engineering examples, one often encounters the need to sample from unnormalized distributions with complex shapes that may also be implicitly defined through a physical or numerical simulation model, making it computationally expensive to evaluate the associated density function. For such cases, MCMC has proven to be an invaluable tool. Random-walk Metropolis Methods (also known as Metropolis-Hastings (MH)), in particular, are highly popular for their simplicity, flexibility, and ease of implementation. However, most MH algorithms suffer from significant limitations when attempting to sample from distributions with multiple modes (particularly disconnected ones). Here, in this paper, we present Intrepid MCMC - a novel MH scheme that utilizes a simple coordinate transformation to significantly improve the mode-finding ability and convergence rate to the target distribution of random-walk Markov chains while retaining most of the simplicity of the vanilla MH paradigm. Through multiple examples, we showcase the improvement in the performance of Intrepid MCMC over vanilla MH for a wide variety of target distribution shapes. We also provide an analysis of the mixing behavior of the Intrepid Markov chain, as well as the efficiency of our algorithm for increasing dimensions. A thorough discussion is presented on the practical implementation of the Intrepid MCMC algorithm. Finally, its utility is highlighted through a Bayesian parameter inference problem for a two-degree-of-freedom oscillator under free vibration.

97 - MATHEMATICS AND COMPUTING↗
DOE OSTI2024

TDCOSMO - XVI. Measurement of the Hubble constant from the lensed quasar WGD 2038–4008

Time-delay cosmography is a powerful technique to constrain cosmological parameters, particularly the Hubble constant (H0). The TDCOSMO Collaboration is performing an ongoing analysis of lensed quasars to constrain cosmology using this method. In this work, we obtain constraints from the lensed quasar WGD 2038−4008 using new time-delay measurements and previous mass models by TDCOSMO. This is the first TDCOSMO lens to incorporate multiple lens modeling codes and the full time-delay covariance matrix into the cosmological inference. The models are fixed before the time delay is measured, and the analysis is performed blinded with respect to the cosmological parameters to prevent unconscious experimenter bias. We obtain DΔ t = 1.68−0.38+0.40 Gpc using two families of mass models, a power-law describing the total mass distribution, and a composite model of baryons and dark matter, although the composite model is disfavored due to kinematics constraints. In a flat ΛCDM cosmology, we constrain the Hubble constant to be H0 = 65−14+23 km s−1 Mpc−1. The dominant source of uncertainty comes from the time delays, due to the low variability of the quasar. Future long-term monitoring, especially in the era of the Vera C. Rubin Observatory’s Legacy Survey of Space and Time, could catch stronger quasar variability and further reduce the uncertainties. This system will be incorporated into an upcoming hierarchical analysis of the entire TDCOSMO sample, and improved time delays and spatially-resolved stellar kinematics could strengthen the constraints from this system in the future.Key words: gravitational lensing: strong / cosmological parameters / distance scale⋆ Corresponding author; kcwong19@gmail.com.⋆⋆ NHFP Einstein fellow.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

Selective Amnesia using Contrastive Subnet Erasure for Class Level Unlearning in Vision Models

We study concept-level forgetting in pretrained vision models: removing an entire semantic category so the system no longer recognizes that object in unseen images and contexts, rather than merely forgetting specific training examples. Prior work either applies blunt global projections or fine-tunes parameters, which can introduce collateral damage to unrelated features, add compute, and become unstable as forgetting strength increases. We introduce Contrastive Subnet Erasure (CSE), a training-free, encoder-centric edit that targets a compact set of channels most responsible for the class and attenuates them in a calibrated manner. The modification is algebraically folded into the subsequent layer, yielding no inference-time overhead and leaving task heads unchanged. To evaluate whether forgetting generalizes beyond the data used to specify the class, we introduce a cross dataset protocol in which the class is defined on a source dataset and performance is measured on a disjoint target dataset drawn from a different distribution with no shared images. This setup tests whether the model still fails to recognize the object when it looks different or appears in new scenes, and it helps avoid overfitting to patterns in the source dataset. Across CIFAR 10, CIFAR 100, and ImageNet under this protocol, CSE achieves stronger forgetting of the target class while better preserving non target utility than existing baselines in both single class and multi class settings. Overall, CSE provides a simple, stable, and deployment-ready mechanism for class-level unlearning in vision.

Kotevska, Olivera [ORNL] (ORCID:0000000316772243)↗
DOE OSTI2025

Simultaneous inference of equation of state parameters and unknown data errors with uncertainty quantification via hierarchical Bayesian posterior maximization

Equations of state (EOSs) are a key component in running hydrodynamic simulations as they relate the thermodynamic states for the material. The Davis reactants EOS is commonly used for modeling high explosives (HEs), and the EOS model parameters are calibrated using material specific data. The calibrations are often performed with uncertainty quantification via Bayesian inference to account for uncertainty in the data and generate ensembles of likely parameters. However, there are relatively few HE data sets to use for calibration and many are historical and lack error information. In this work, we simultaneously calibrate the Davis reactants EOS model parameters and unknown data error terms for the high explosive PBX 9501. To quantify the uncertainty in the models and the data, we use a Bayesian framework for the calibration and compute the hierarchical Bayesian posterior distribution with both a posteriori maximization approach and Markov Chain Monte Carlo. In general, we find that, given our assumptions, the two approaches result in similar calibrated parameters, posterior covariance matrices, and insights about the parameters but that the posterior maximization requires far less computational resources.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Topographical Controls on Hillslope-Scale Hydrology Drive Shrub Distributions on the Seward Peninsula, Alaska

Observations indicate shrubs are expanding across the Arctic tundra, mainly on hillslopes and primarily in response to climate warming. However, the impact topography exerts on hydrology, nutrient dynamics, and plant growth can make untangling the mechanisms behind shrub expansion difficult. We examined the role topography plays in determining shrub expansion by applying a coupled transect version of a mechanistic ecosystem model (ecosys) in a tundra hillslope site in the Seward Peninsula, Alaska. Modeled biomass of the dominant plant functional types agreed well with field measurements (R 2 = 0.89) and accurately represented shrub expansion over the past 30 years inferred from satellite observations. In the well-drained crest position, canopy water potential and plant nitrogen (N) uptake was modeled to be low from plant and microbial water stress. Intermediate soil water content in the mid-slope position enhanced mineralization and plant N uptake, increasing shrub biomass. The deciduous shrub growth in the mid-slope position was further enhanced by symbiotic N 2 fixation primed by increased root carbon allocation. The gentle slope in the poorly drained lower-slope position resulted in saturated soil conditions that reduced soil O 2 concentrations, leading to lower root O 2 uptake and lower nutrient uptake and plant biomass. Additionally, a simulation that removed topographical interconnectivity between grid cells resulted in (1) a 28% underestimate of mean shrub biomass and (2) over or underestimated shrub productivity at the various hillslope positions. Our results indicate that land models need to account for hillslope-scale coupled surface and subsurface hydrology to accurately predict current plant distributions and future trajectories in Arctic ecosystems.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

A genetic algorithm approach for reconstructing spectral content from filtered x-ray diode array spectrometers

Filtered diode array spectrometers are routinely employed to infer the temporal evolution of spectral power from x-ray sources, but uniquely extracting spectral content from a finite set of broad, spectrally overlapping channel spectral sensitivities is decidedly nontrivial in these under-determined systems. Here, we present the use of genetic algorithms to reconstruct a probabilistic spectral intensity distribution and compare to the traditional approach most commonly found in the literature. Unlike many of the previously published models, spectral reconstructions from this approach are neither limited by basis functional forms nor do they require a priori spectral knowledge. While the original intent of such measurements was to diagnose the temporal evolution of spectral power from quasi-blackbody radiation sources—where the exact details of spectral content were not thought to be crucial—we demonstrate that this new technique can greatly enhance the utility of the diagnostic by providing more physical spectra and improved robustness to hardware configuration for even strongly non-Planckian distributions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

Whole-genome sequencing distinguishes the two most common giant kelp ecomorphs

Abstract Giant kelp, Macrocystis pyrifera, exists as distinct morphological variants—or “ecomorphs”—in different populations, yet the mechanism for this variation is uncertain, and environmental drivers for either adaptive or plastic phenotypes have not been identified. The ecomorphs Macrocystis “pyrifera” and M. “integrifolia” are distributed throughout temperate waters of North and South America with almost no geographic overlap and exhibit an incongruous, non-mirrored, distribution across the equator. This study evaluates the degree of genetic divergence between M. “pyrifera” and M. “integrifolia” across 18 populations in Chile and California using whole-genome sequencing and single-nucleotide polymorphism markers. Our results based on a principal component analysis, admixture clustering by genetic similarity, and phylogenetic inference demonstrate that M. “pyrifera” and M. “integrifolia” are genetically distinguishable. Analyses reveal separation by Northern and Southern Hemispheres and between morphs within hemispheres, suggesting that the convergent “integrifolia” morphology arose separately in each hemisphere. This is the first study to use whole-genome sequencing to understand genetic divergence in giant kelp ecomorphs, identifying 83 potential genes under selection and providing novel insights about Macrocystis evolution that were not evident with previous genetic techniques. Future studies are needed to uncover the environmental forces driving local adaptation and presumed convergent evolution of these morphs.

Environmental Sciences & Ecology↗
DOE OSTI2021

Bayesian estimations of orientation distribution functions from small-angle scattering enable direct prediction of mechanical stress in anisotropic materials

Properties of soft materials are influenced by their anisotropic structuring under nonequilibrium fields. Although anisotropic structure-property relationships have been extensively explored theoretically, comparison to experiments requires determination of the microstructural orientation probability distribution function (OPDF) of microstructural elements. Small angle scattering (SAS) measurements encode information about the OPDF, but tools to navigate this connection are incomplete. Here, we develop and validate an explicit framework to link arbitrary OPDFs to SAS measurements. Specifically, we propose, validate, and apply a method, maximum a posteriori scattering inference (MAPSI), whereby the OPDF may be obtained from SAS measurements using a Bayesian estimation method. Using this method, we obtain estimates of the full 3D OPDF for two model semidilute fd-virus (rodlike) dispersions at concentrations that are approximately equal to and twice the overlap concentration. From the OPDF, we calculate its second and fourth moments and compare these to predictions for a dilute suspension of rigid rods and to a recent theory for semidilute suspensions. Finally, we use both the theoretical and measured moments to calculate the stress, both for dilute and semidilute suspensions. These predictions are not only compared to each other, but also to measured values of the shear stress, and point to new insights into the behavior of suspensions of highly elongated particles in the transition between dilute and semidilute behavior. We also use this new framework to provide perspective on the connection between scalar parameterizations of scattering and the OPDF that have frequently been used in the past. Here, the new tools developed in this work provide an unprecedented path toward experimental validation of dynamical theories of rodlike colloids and polymers, and for measurement of nonequilibrium structures and stresses of other complex fluids and soft materials with SAS.

36 MATERIALS SCIENCE↗