Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “applied statistics”

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 793 records · Page 44
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Geometric frustration in the myosin superlattice of vertebrate muscle

Geometric frustration results from an incompatibility between minimum energy arrangements and the geometry of a system, and gives rise to interesting and novel phenomena. Here, we report geometric frustration in a native biological macromolecular system---vertebrate muscle. We analyse the disorder in the myosin filament rotations in the myofibrils of vertebrate striated (skeletal and cardiac) muscle, as seen in thin-section electron micrographs, and show that the distribution of rotations corresponds to an archetypical geometrically frustrated system---the triangular Ising antiferromagnet. Spatial correlations are evident out to at least six lattice spacings. The results demonstrate that geometric frustration can drive the development of structure in complex biological systems, and may have implications for the nature of the actin--myosin interactions involved in muscle contraction. Identification of the distribution of myosin filament rotations with an Ising model allows the extensive results on the latter to be applied to this system. It shows how local interactions (between adjacent myosin filaments) can determine long-range order and, conversely, how observations of long-range order (such as patterns seen in electron micrographs) can be used to estimate the energetics of these local interactions. Furthermore, since diffraction by a disordered system is a function of the second-order statistics, the derived correlations allow more accurate diffraction calculations, which can aid in interpretation of X-ray diffraction data from muscle specimens for structural analysis.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Constraining the Twomey effect from satellite observations: issues and perspectives

Abstract. The Twomey effect describes the radiative forcing associated with a change in cloud albedo due to an increase in anthropogenic aerosol emissions. It is driven by the perturbation in cloud droplet number concentration (ΔNd, ant) in liquid-water clouds and is currently understood to exert a cooling effect on climate. The Twomey effect is the key driver in the effective radiative forcing due to aerosol–cloud interactions, but rapid adjustments also contribute. These adjustments are essentially the responses of cloud fraction and liquid water path to ΔNd, ant and thus scale approximately with it. While the fundamental physics of the influence of added aerosol particles on the droplet concentration (Nd) is well described by established theory at the particle scale (micrometres), how this relationship is expressed at the large-scale (hundreds of kilometres) perturbation, ΔNd, ant, remains uncertain. The discrepancy between process understanding at particle scale and insufficient quantification at the climate-relevant large scale is caused by co-variability of aerosol particles and updraught velocity and by droplet sink processes. These operate at scales on the order of tens of metres at which only localised observations are available and at which no approach yet exists to quantify the anthropogenic perturbation. Different atmospheric models suggest diverse magnitudes of the Twomey effect even when applying the same anthropogenic aerosol emission perturbation. Thus, observational data are needed to quantify and constrain the Twomey effect. At the global scale, this means satellite data. There are four key uncertainties in determining ΔNd, ant, namely the quantification of (i) the cloud-active aerosol – the cloud condensation nuclei (CCN) concentrations at or above cloud base, (ii) Nd, (iii) the statistical approach for inferring the sensitivity of Nd to aerosol particles from the satellite data and (iv) uncertainty in the anthropogenic perturbation to CCN concentrations, which is not easily accessible from observational data. This review discusses deficiencies of current approaches for the different aspects of the problem and proposes several ways forward: in terms of CCN, retrievals of optical quantities such as aerosol optical depth suffer from a lack of vertical resolution, size and hygroscopicity information, non-direct relation to the concentration of aerosols, difficulty to quantify it within or below clouds, and the problem of insufficient sensitivity at low concentrations, in addition to retrieval errors. A future path forward can include utilising co-located polarimeter and lidar instruments, ideally including high-spectral-resolution lidar capability at two wavelengths to maximise vertically resolved size distribution information content. In terms of Nd, a key problem is the lack of operational retrievals of this quantity and the inaccuracy of the retrieval especially in broken-cloud regimes. As for the Nd-to-CCN sensitivity, key issues are the updraught distributions and the role of Nd sink processes, for which empirical assessments for specific cloud regimes are currently the best solutions. These considerations point to the conclusion that past studies using existing approaches have likely underestimated the true sensitivity and, thus, the radiative forcing due to the Twomey effect.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Robust global detection of forced changes in mean and extreme precipitation despite observational disagreement on the magnitude of change

Detection and attribution (D&A) of forced precipitation change are challenging due to internal variability, limited spatial, and temporal coverage of observational records and model uncertainty. These factors result in a low signal-to-noise ratio of potential regional and even global trends. Here, we use a statistical method – ridge regression – to create physically interpretable fingerprints for the detection of forced changes in mean and extreme precipitation with a high signal-to-noise ratio. The fingerprints are constructed using Coupled Model Intercomparison Project phase 6 (CMIP6) multi-model output masked to match coverage of three gridded precipitation observational datasets – GHCNDEX, HadEX3, and GPCC – and are then applied to these observational datasets to assess the degree of forced change detectable in the real-world climate in the period 1951–2020. We show that the signature of forced change is detected in all three observational datasets for global metrics of mean and extreme precipitation. Forced changes are still detectable from changes in the spatial patterns of precipitation even if the global mean trend is removed from the data. This shows the detection of forced change in mean and extreme precipitation beyond a global mean trend is robust and increases confidence in the detection method's power as well as in climate models' ability to capture the relevant processes that contribute to large-scale patterns of change. We also find, however, that detectability depends on the observational dataset used. Not only coverage differences but also observational uncertainty contribute to dataset disagreement, exemplified by the times of emergence of forced change from internal variability ranging from 1998 to 2004 among datasets. Furthermore, different choices for the period over which the forced trend is computed result in different levels of agreement between observations and model projections. These sensitivities may explain apparent contradictions in recent studies on whether models under- or overestimate the observed forced increase in mean and extreme precipitation. Lastly, the detection fingerprints are found to rely primarily on the signal in the extratropical Northern Hemisphere, which is at least partly due to observational coverage but potentially also due to the presence of a more robust signal in the Northern Hemisphere in general.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Sampling Size Optimization for Bioburden Density Estimation in Planetary Protection

Planetary protection (PP) is a discipline that focuses on minimizing the biological contamination of spacecraft to ensure compliance with international policy. Precise estimation of bioburden - the total number of microbes in or on spacecraft hardware – and the bioburden density are of utmost importance for PP. Such estimation is the way concordance with requirements is demonstrated, and it is critical for quantifying the potential risk of inadvertently contaminating other planetary bodies. Although a suite of molecular techniques have been used to thoroughly characterize and profile the microbiome of various cleanroom environments and spacecraft, the gold standard remains the physical enumeration of microbes via culturing of samples directly taken from spacecraft and associated surfaces. However, due to technical, budgetary, and programmatic constraints, only a manageable portion (around 10%) of the entire spacecraft surface is directly sampled with cotton swabs or wipes. To generate the bioburden current best estimate (CBE) for components not directly verifiable, the accepted approach is to apply a NASA-defined bioburden estimate based on the components’ manufacturing or assembly environment. This approach utilizes a prespecified bioburden density estimation that applies a maximum value across the total surface area of the specified component. For hardware components that underwent similar assembly processes, an implied bioburden is adopted for all components, based on a direct verification of a representative component within the same lot. Once all components have a CBE, the bioburden estimates are generated. In previous publication [ 1], we have shown that statistical risks quantifying the accuracy of the estimates for sampled, prespecified, and implied components can be derived and ranked. For mean squared error (MSE) function, the risks are available analytically and hence a cost function can be obtained to optimize the risks with respect to the sampling area and sampling cost. Since the sampling area and sampling cost are two complimentary variables, their sum will have a well-defined minimum. This paper presents the multivariate optimization of the integrated risk of an empirical Bayes estimator to determine the optimal sampling schedule for a given number of components. It is assumed that given a number of components, N, the bioburden density for each component can either be sampled, implied, or prespecified. The multivariate optimization searches through different options to sample, imply or prespecify the bioburden density for a component, and account for the component’s surface area and cost of sampling. The idea of the optimization is based on the observation that the statistical risk of using an estimator is a monotonically decreasing function of the sampled area. The larger the sampled area, the lower the risk of using the estimator as the estimator becomes more and more accurate as the sampling area increases. On the other hand, the cost of sampling is monotonically increasing as the sampled surface grows. This makes the risk and total cost of sampling complimentary variables which can be counterbalanced to achieve an optimal overall value with respect to the sampled surface. In this paper, the integrated risk has been used to quantify the accuracy of the estimator. This risk has been selected because it depends on neither the true value of the parameter nor on the collected data. The cost of each sample was also available to obtain the total cost of sampling of N components. The paper will present the results based on computer-simulated data as well as the data collected during the InSight mission. The computer-simulated data have N components with randomly generated total areas and each component assigned to one of the three categories according to the method of estimating of bioburden density: sampled, implied, or prespecified. The cost of sampling is also available. The cost of sampling is estimated based on a cost model provided by the planetary protection group at JPL. For this paper, the overall cost was assumed to be a linear function of exposure. The optimization process finds the allocation of the components to the three categories that minimizes the tradeoff between integrated risk and total cost. For the InSight data, a set of components is selected representing all three categories, and optimization is performed to determine if the performed allocation was optimal or if a better allocation could have been obtained. To the best of our knowledge, this work is the first attempt not only perform an accurate estimation of bioburden density but also do it in an optimal way.

97 - MATHEMATICS AND COMPUTING↗
DOE OSTI2024

A principle of maximum ignorance for semiclassical gravity

The principle of maximum ignorance posits that the coarse-grained description of a system is maximally agnostic about its underlying microscopic structure. We briefly review this principle for random matrix theory and for the eigenstate thermalization hypothesis. We then apply this principle in holography to construct ensembles of random mixed states. This leads to an ensemble of microstates which models our microscopic ignorance, and which on average reproduces the effective semiclassical physics of a given bulk state. We call this ensemble the state-averaging ansatz. The output of our model is a prediction for semiclassical contributions to variances and higher statistical moments over the ensemble of microstates. The statistical moments provide coarse-grained — yet gravitationally non-perturbative — information about the microstructure of the individual states of the ensemble. We show that these contributions exactly match the on-shell action of known wormhole configurations of the gravitational path integral. These results strengthen the view that wormholes simply parametrize the ignorance of the microstructure of a fundamental state, given a fixed semiclassical bulk description.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Acceleration of Solvation Free Energy Calculation via Thermodynamic Integration Coupled with Gaussian Process Regression and Improved Gelman–Rubin Convergence Diagnostics

The determination of the solvation free energy of ions and molecules holds profound importance across a spectrum of applications spanning chemistry, biology, energy storage, and the environment. Molecular dynamics simulations are powerful tools for computing this critical parameter. Nevertheless, the accurate and efficient calculation of the solvation free energy becomes a formidable endeavor when dealing with complex systems characterized by potent Coulombic interactions and sluggish ion dynamics and, consequently, slow transition across various metastable states. Here, in the present study, we expose limitations stemming from the conventional calculation of the statistical inefficiency g in the thermodynamic integration method, a factor that can hinder the determination of convergence of the solvation free energy and its associated uncertainty. Instead, we propose a robust scheme based on Gelman–Rubin convergence diagnostics. We leverage this improved estimation of uncertainties to introduce an innovative accelerated thermodynamic integration method based on the Gaussian Process regression. This methodology is applied to the calculation of the solvation free energy of trivalent rare-earth elements immersed in ionic liquids, a scenario in which the aforementioned challenges render standard approaches ineffective. The proposed method proves to be effective in computing solvation free energy in situations where traditional thermodynamic integration methods fall short.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Evaluating the location capabilities of a regional infrasonic network in Utah, US, using both ray tracing-derived and empirical-derived celerity-range and backazimuth models

SUMMARY More realistic models for infrasound signal propagation across a region can be used to improve the precision and accuracy of spatial and temporal source localization estimates. Motivated by incomplete infrasound event bulletins in the Western US, the location capabilities of a regional infrasonic network of stations located between 84–458 km from the Utah Test and Training Range, Utah, USA, is assessed using a series of near-surface explosive events with complementary ground truth (GT) information. Signal arrival times and backazimuth estimates are determined with an automatic F-statistic based signal detector and manually refined by an analyst. This study represents the first application of three distinct celerity-range and backazimuth models to an extensive suite of realistic signal detections for event location purposes. A singular celerity and backazimuth deviation model was previously constructed using ray tracing analysis based on an extensive archive of historical atmospheric specifications and is applied within this study to test location capabilities. Similarly, a set of multivariate, season and location specific models for celerity and backazimuth are compared to an empirical model that depends on the observations across the infrasound network and the GT events, which accounts for atmospheric propagation variations from source to receiver. Discrepancies between observed and predicted signal celerities result in locations with poor accuracy. Application of the empirical model improves both spatial localization precision and accuracy; all but one location estimates retain the true GT location within the 90 per cent confidence bounds. Average mislocation of the events is 15.49 km and average 90 per cent error ellipse areas are 4141 km2. The empirical model additionally reduces origin time residuals; origin time residuals from the other location models are in excess of 160 s while residuals produced with the empirical model are within 30 s of the true origin time. We demonstrate that event location accuracy is driven by a combination of signal propagation model and the azimuthal gap of detecting stations. A direct relationship between mislocation, error ellipse area and increased station azimuthal gaps indicate that for sparse networks, detection backazimuths may drive location biases over traveltime estimates.

58 GEOSCIENCES↗
DOE OSTI2024

Estimated impacts of forest restoration scenarios on smoke exposures among outdoor agricultural workers in California

Abstract As wildfires continue to worsen across western United States, forest managers are increasingly employing prescribed burns as a way to reduce excess fuels and future wildfire risk. While the ecological benefits of these fuel treatments are clear, little is known about the smoke exposure tradeoffs of using prescribed burns to mitigate wildfires, particularly among at-risk populations. Outdoor agricultural workers are a population at increased risk of smoke exposure because of their time spent outside and the physical demands of their work. Here, we assess the smoke exposure impacts among outdoor agricultural workers resulting from the implementation of six forest management scenarios proposed for a landscape in the Central Sierra, California. We leverage emissions estimates from LANDIS-II to model daily PM 2.5 concentrations with the Hybrid Single Particle Lagrangian Integrated Trajectory Model (HYSPLIT) and link those to agricultural employment data from the Bureau of Labor Statistics. We find a u-shaped result, in that moderate amounts of prescribed burning result in the greatest reduction in total smoke exposure among outdoor agricultural workers, particularly during months of peak agricultural activity due to wildfire-specific smoke reductions. The reduction in total smoke exposure, relative to scenarios with the least amount of management, decreases as more prescribed burning is applied to the landscape due to the contributions of the fuel treatments themselves to overall smoke burden. The results of this analysis may contribute to preparedness efforts aimed at reducing smoke exposures among outdoor agricultural workers, while also informing forest management planning for this specific landscape.

Environmental Sciences & Ecology↗
DOE OSTI2023

The satellite population around luminous red galaxies in the 25 deg2 DESI Legacy Imaging Surveys Early Data Release

ABSTRACT Luminous red galaxies, or LRGs, are representative of the most massive galaxies and were originally selected in the Sloan Digital Sky Survey as good tracers of large-scale structure. They are dominated by by uniformly old stellar populations, have low star formation rates, early type morphologies, and little cold gas. Despite having old stellar populations and little in situ star formation, studies have shown that they have grown their stellar mass since z = 1, implying that they grow predominantly via the accretion of satellites. Tests of this picture have been limited because of the lack of deep imaging data sets that both covers a large enough area of the sky to contain substantial numbers of LRGs and that also is deep enough to detect faint satellites. We use the 25 deg2 Early Data Release (EDR) of the DESI Legacy Imaging Surveys to characterize the satellite galaxy population of LRGs out to z = 0.65. The DESI Legacy Imaging Surveys are comprised of grz imaging to 2–2.5 mag deeper than SDSS and with better image quality. We use a new statistical background technique to identify excess populations of putative satellite galaxies around 1823 LRGs at 0.2 < z < 0.65. In three redshift and luminosity bins we measure the numbers of satellite galaxies and their r−z colour distribution down to rest-frame g-band luminosity limits at least 3.6 times fainter than L*. In addition, we develop a forward modeling technique and apply it to constrain the mean number of satellites in each of our redshift and luminosity bins. Finally, we use these estimates to determine the amount of stellar mass growth in LRGs down to the local Universe.

Astronomy & Astrophysics↗
DOE OSTI2025

Block Lanczos algorithm for lattice QCD spectroscopy and matrix elements

Recent work introduced a new framework for analyzing correlation functions with improved convergence and signal-to-noise properties, as well as rigorous quantification of excited-state effects, based on the Lanczos algorithm and spurious eigenvalue filtering with the Cullum-Willoughby test. Here, we extend this framework to the analysis of correlation-function matrices built from multiple interpolating operators in lattice quantum chromodynamics (QCD) by constructing an oblique generalization of the block Lanczos algorithm, as well as a new physically motivated reformulation of the Cullum-Willoughby test that generalizes to block Lanczos straightforwardly. The resulting block Lanczos method directly extends generalized eigenvalue problem (GEVP) methods, which can be viewed as applying a single iteration of block Lanczos. Block Lanczos provides qualitative and quantitative advantages over GEVP methods analogous to the benefits of Lanczos over the standard effective mass, including faster convergence to ground- and excited-state energies, explicitly computable two-sided error bounds, straightforward extraction of matrix elements of external currents, and asymptotically constant signal-to-noise. No fits or statistical inference are required. Proof-of-principle calculations are performed for noiseless mock-data examples as well as two-by-two proton correlation-function matrices in lattice QCD.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Optimizing fictitious states for Bell inequality violation in bipartite qubit systems with applications to the t t ¯ system

There is a significant interest in testing quantum entanglement and Bell inequality violation in high-energy experiments. Since the analyses in high-energy experiments are performed with events statistically averaged over phase space, the states used to determine observables depend on the choice of coordinates through an event-dependent basis and are thus not genuine quantum states, but rather “fictitious states.” We find that the basis which diagonalizes the spin-spin correlations is optimal for constructing fictitious states to test the violation of Bell’s inequality. This result is applied directly to the bipartite qubit system of a top and antitop produced at a hadron collider. We show that the beam axis is the optimal basis choice near the t t ¯ threshold production for measuring Bell inequality violation, while at high transverse momentum the basis that aligns along the momentum direction of the top is optimal. Published by the American Physical Society 2024

Astronomy & Astrophysics↗
DOE OSTI2023

Enhanced Hanford Low-Activity Waste Glass Property Data Development: Phase 5 and Phase 6

This report summarizes the data collected on two test matrices of low-activity waste (LAW) glass compositions intended to expand the composition-property database: Low-Activity Waste (LAW) Phase 5: Expansion of LAW Glass Composition Boundaries and LAW Phase 6: High PCT and VHT Response Glass Matrix. Both matrix glass compositions were statistically designed to expand the LAW glass composition region. The analyses performed on these glasses include chemical composition (for target compositional verification), density, viscosity, electrical conductivity, crystal fraction, container centerline cooling with crystal identification, the product consistency test (PCT) response, the vapor hydration test (VHT) response, and sulfur solubility. Because of the slightly different scope of the two matrices, not all methods were applied to both matrices. Specifically, the following measurements were taken only on the LAW Phase 5: Expansion of LAW Glass Composition Boundaries glasses: crystal fraction as a function of temperature, density (ρ), viscosity (η), and electrical conductivity (EC, e). Combined, these data contribute a significant amount, 51 glasses, to the database for high LAW loaded enhanced waste glasses. Most of these data are focused near the boundaries of acceptable PCT and VHT responses, where prediction uncertainties are most impactful.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

Domain Adaptive Graph Neural Networks for Constraining Cosmological Parameters Across Multiple Data Sets

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross-dataset tasks. Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Bridging the Gap between Cosmological Simulations with Graph Neural Networks and Domain Adaptation

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross dataset tasks (up to 28% better relative error and up to almost an order of magnitude better χ 2 ). Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Understanding Type Ia Supernova Distance Biases by Simulating Spectral Variations

In the next decade, transient searches from the Vera C. Rubin Observatory and the Nancy Grace Roman Space Telescope will increase the sample of known Type Ia supernovae (SNe Ia) from ~10 3 to 10 5 . With this reduction of statistical uncertainties on cosmological measurements, new methods are needed to reduce systematic uncertainties. Here, characterizing the underlying spectroscopic evolution of SN Ia remains a major systematic uncertainty in current cosmological analyses, motivating a new simulation tool for the next era of SN Ia cosmology: Build Your Own Spectral Energy Distribution (byosed). byosed is used within the SNANA framework to simulate light curves by applying spectral variations to model SEDs, enabling flexible testing of possible systematic shifts in SN Ia distance measurements. We test the framework by comparing a nominal Roman SN Ia survey simulation using a baseline SED model to simulations using SEDs perturbed with byosed, and investigating the impact of ignoring specific SED features in the analysis. These features include semiempirical models of two possible, predicted relationships: between SN ejecta velocity and light-curve observables, and a redshift-dependent relationship between SN Hubble residuals and host-galaxy mass. We analyze each byosed simulation using the SALT2 and BEAMS with Bias Corrections framework, and estimate changes in the measured value of the dark-energy equation-of-state parameter, w. We find a difference of Δw = –0.023 for SN velocity and Δw = 0.021 for redshift-evolving host mass when compared to simulations without these features. By using byosed for SN Ia cosmology simulations, future analyses (e.g., the Rubin and Roman SN Ia samples) will have greater flexibility to constrain or reduce such SN Ia modeling uncertainties.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Reconstructing Ly$α$ Fields from Low-resolution Hydrodynamical Simulations with Deep Learning

Hydrodynamical cosmological simulations are a powerful tool for accurately predicting the properties of the intergalactic medium (IGM) and for producing mock skies that can be compared against observational data. However, the need to resolve density fluctuation in the IGM puts a stringent requirement on the resolution of such simulations, which in turn limits the volumes that can be modeled, even on the most powerful supercomputers. In this work, we present a novel modeling method that combines physics-driven simulations with data-driven generative neural networks to produce outputs that are qualitatively and statistically close to the outputs of hydrodynamical simulations employing eight times higher resolution. We show that the Ly$α$ flux field, as well as the underlying hydrodynamic fields, have greatly improved statistical fidelity over a low-resolution simulation. Importantly, the design of our neural network allows for sampling multiple realizations from a given input, enabling us to quantify the model uncertainty. Using test data, we demonstrate that this model uncertainty correlates well with the true error of the Ly$α$ flux prediction. Ultimately, our approach allows for training on small simulation volumes and applying it to much larger ones, opening the door to producing accurate Ly$α$ mock skies in volumes of Hubble size, as will be probed with DESI and future spectroscopic sky surveys.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Domain Adaptive Graph Neural Networks for Constraining Cosmological Parameters Across Multiple Data Sets

Deep learning models have been shown to outperform methods that rely on summary statistics, like the power spectrum, in extracting information from complex cosmological data sets. However, due to differences in the subgrid physics implementation and numerical approximations across different simulation suites, models trained on data from one cosmological simulation show a drop in performance when tested on another. Similarly, models trained on any of the simulations would also likely experience a drop in performance when applied to observational data. Training on data from two different suites of the CAMELS hydrodynamic cosmological simulations, we examine the generalization capabilities of Domain Adaptive Graph Neural Networks (DA-GNNs). By utilizing GNNs, we capitalize on their capacity to capture structured scale-free cosmological information from galaxy distributions. Moreover, by including unsupervised domain adaptation via Maximum Mean Discrepancy (MMD), we enable our models to extract domain-invariant features. We demonstrate that DA-GNN achieves higher accuracy and robustness on cross-dataset tasks (up to $28\%$ better relative error and up to almost an order of magnitude better $\chi^2$). Using data visualizations, we show the effects of domain adaptation on proper latent space data alignment. This shows that DA-GNNs are a promising method for extracting domain-independent cosmological information, a vital step toward robust deep learning for real cosmic survey data.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Hyper Spectral Anomaly Detection

Anomaly detection is a common machine learning (ML) task with growing importance in the fields of imaging, quality assurance, and multiple security related disciplines. Anomaly detection is more difficult than traditional machine learning methods due to the inherent unlabeled nature of the datasets. Existing anomaly detection architectures commonly face challenges with explainability, retaining information related to the relational structure of the data, and false positive rates. Hyperspectral Imaging Anomaly Detection (HSI) is a statistical model that employs vertex and edge weighted graphs to preserve the data’s relationships on different topographical scales. The model is able to generalize from anomaly detection in 2D images to novel datasets related to cyber-security. Furthermore, the use of multi-spectral and other filtering methods results in fewer false positives and increases the explainability of model predictions. When applying HSI to cyber-security datasets, we are able to successfully detect malicious activity with a relatively high degree of accuracy.

97 - MATHEMATICS AND COMPUTING↗