Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Bayesian 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.

1,237 records · Page 69

ON THE MASS DISTRIBUTION AND BIRTH MASSES OF NEUTRON STARS

We investigate the distribution of neutron star masses in different populations of binaries, employing Bayesian statistical techniques. In particular, we explore the differences in neutron star masses between sources that have experienced distinct evolutionary paths and accretion episodes. We find that the distribution of neutron star masses in non-recycled eclipsing high-mass binaries as well as of slow pulsars, which are all believed to be near their birth masses, has a mean of 1.28 M{sub Sun} and a dispersion of 0.24 M{sub Sun }. These values are consistent with expectations for neutron star formation in core-collapse supernovae. On the other hand, double neutron stars, which are also believed to be near their birth masses, have a much narrower mass distribution, peaking at 1.33 M{sub Sun }, but with a dispersion of only 0.05 M{sub Sun }. Such a small dispersion cannot easily be understood and perhaps points to a particular and rare formation channel. The mass distribution of neutron stars that have been recycled has a mean of 1.48 M{sub Sun} and a dispersion of 0.2 M{sub Sun }, consistent with the expectation that they have experienced extended mass accretion episodes. The fact that only a very small fraction of recycled neutron stars in the inferred distribution have masses that exceed {approx}2 M{sub Sun} suggests that only a few of these neutron stars cross the mass threshold to form low-mass black holes.

79 ASTRONOMY AND ASTROPHYSICS↗

Learning functional priors and posteriors from data and physics

In this work, we develop a new Bayesian framework based on deep neural networks to be able to extrapolate in space-time using historical data and to quantify uncertainties arising from both noisy and gappy data in physical problems. Specifically, the proposed approach has two stages: (1) prior learning and (2) posterior estimation. At the first stage, we employ the physics-informed Generative Adversarial Networks (PI-GAN) to learn a functional prior either from a prescribed function distribution, e.g., Gaussian process, or from historical data and physics. At the second stage, we employ the Hamiltonian Monte Carlo (HMC) method to estimate the posterior in the latent space of PI-GANs. In addition, we use two different approaches to encode the physics: (1) automatic differentiation, used in the physicsinformed neural networks (PINNs) for scenarios with explicitly known partial differential equations (PDEs), and (2) operator regression using the deep operator network (DeepONet) for PDE-agnostic scenarios. We then test the proposed method for (1) meta-learning for one-dimensional regression, and forward/inverse PDE problems (combined with PINNs); (2) PDE-agnostic physical problems (combined with DeepONet), e.g., fractional diffusion as well as saturated stochastic (100-dimensional) flows in heterogeneous porous media; and (3) spatial-temporal regression problems, i.e., inference of a marine riser displacement field using experimental data from the Norwegian Deepwater Programme (NDP). The results demonstrate that the proposed approach can provide accurate predictions as well as uncertainty quantification given very limited scattered and noisy data, since historical data could be available to provide informative priors. In summary, the proposed method is capable of learning flexible functional priors, e.g., both Gaussian and non-Gaussian process, and can be readily extended to big data problems by enabling mini-batch training using stochastic HMC or normalizing flows since the latent space is generally characterized as low dimensional.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Efficient Optimization of Plasma Radiation Detector Configurations using Imperfect Inference Models

The configurations of instruments fielded on an experiment affect the amount of information captured and the quality of subsequent inference. Here, we investigate the problem of optimizing plasma x-ray radiation detectors in a magneto-inertial fusion experiment at Sandia National Laboratories. It is impossible to directly measure properties such as the temperature of the thermonuclear fusion plasma produced in these experiments because of the extreme environment and destructive nature of the experiment. Among other diagnostics, several detectors are placed with significant standoff from the fusion target to capture the x-rays emitted by the fusion plasma, which can be used to infer some of its properties. To optimize the configuration of these detectors, a high-fidelity model (HFM) is used for simulating outputs and a low-fidelity model (LFM) is used for inference. We develop methods based on A- and L-optimality criteria that are efficient to compute while explicitly accounting for the discrepancy between the HFM and the LFM. The method allows us to find detector configurations that perform similarly to or better than the configuration obtained using an existing sampling-based optimization method while decreasing computational time by a factor of 50. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Bayesian optimization↗

LABQ3: Bayesian method for quantification of mineral compositions and nano-scale elemental mapping of 3D synchrotron XCT data

Quantitative analysis of mineral compositions is essential in understanding geochemical, mineralogical and environmental processes. Fine-resolution 3D imaging is widely done using synchrotron X-ray computed tomography (XCT), but existing analyses are limited to visualization and segmentation. This paper presents a new method, Linear Attenuation Bayesian Quantitative 3D-mapper (LABQ3), based on the linearity of X-ray attenuation with respect to elemental concentrations. To address the random variability in attenuation measurements, LABQ3 employs Bayesian decision theory to minimize classification error, using reference attenuation distributions from scans of pure mineral standards. To demonstrate LABQ3 and test its performance, we studied precipitated carbonate samples. XCT scans were done at multiple energies using the transmission X-ray microscope (TXM) at beamline 32-ID-C of the Advanced Photon Source at Argonne National Laboratory. The reconstructed 3D images have a voxel size of 20 nm. Analyses revealed rich nano-scale compositional heterogeneity within individual particles. A mixture of calcium and cadmium produced an overall stoichiometric composition of (Ca 0.78 ,Cd 0.22 )CO 3 , with some voxels containing nearly pure CdCO 3 . The addition of zinc led to an overall stoichiometric composition of 33% Ca, 28% Cd, 39% Zn, with a nearly pure CaCO 3 core and compositional zonation through the rim. These compositional gradients are related to temporal sequences of carbonate mineral formation where Cd precipitated at the beginning in (Ca,Cd)CO 3 , while Cd and Zn precipitated at the end in (Ca, Cd,Zn)CO 3 . Results differ from bulk analyses using Inductively Coupled Plasma-Mass Spectrometry (ICP-MS), showing that LABQ3 provides particle-specific insights. LABQ3 distinguishes itself by quantifying chemical compositions along a continuum, making it different from XCT analyses based on segmentation. LABQ3 allows simultaneous acquisition of morphology and chemical composition in 3D, facilitating the interpretation of chemical gradients of trace elements, quantification of solid solution compositions, inferences about temporal sequences of mineral precipitation, and addressing other concerns about solid-phase chemistry.

58 GEOSCIENCES↗

Scaling and merging time-resolved pink-beam diffraction with variational inference

Time-resolved x-ray crystallography (TR-X) at synchrotrons and free electron lasers is a promising technique for recording dynamics of molecules at atomic resolution. While experimental methods for TR-X have proliferated and matured, data analysis is often difficult. Extracting small, time-dependent changes in signal is frequently a bottleneck for practitioners. Recent work demonstrated this challenge can be addressed when merging redundant observations by a statistical technique known as variational inference (VI). However, the variational approach to time-resolved data analysis requires identification of successful hyperparameters in order to optimally extract signal. In this case study, we present a successful application of VI to time-resolved changes in an enzyme, DJ-1, upon mixing with a substrate molecule, methylglyoxal. We present a strategy to extract high signal-to-noise changes in electron density from these data. Furthermore, we conduct an ablation study, in which we systematically remove one hyperparameter at a time to demonstrate the impact of each hyperparameter choice on the success of our model. We expect this case study will serve as a practical example for how others may deploy VI in order to analyze their time-resolved diffraction data.

47 OTHER INSTRUMENTATION↗

Alleviating prior dependencies for DESI DR1 clustering fits through reparameterization

Bayesian analyses of the full-shape clustering of Dark Energy Spectroscopic Instrument (DESI) Data Release 1 (DR1) exhibit prior-volume projection effects, whereby weakly constrained nuisance parameters of the Effective Field Theory of Large Scale Structure (EFTofLSS) shift marginalized cosmological posteriors away from the posterior maximum. We reanalyze DESI DR1 power spectrum multipoles using two complementary mitigation strategies: (i) nonlinear orthogonalization to decorrelate nuisance and cosmological parameter priors, and (ii) a fully reparameterization-invariant Jeffreys prior over all EFTofLSS coefficients, evaluated on-the-fly via closed-form Jacobians. Including data from DESI, Big-Bang Nuclesynthesis and a constraint on $n_{\mathrm{s}}$, baseline priors lead to multi-$σ$ projection in the Hubble parameter $H_{0}$ and dark energy equation of state parameters $w_{0}$ and $w_{a}$; the Jeffreys prior successfully recenters these posteriors to enclose the maximum a posteriori estimate within the 68% credible regions, demonstrating clear mitigation of projection effects for these late-time expansion parameters. A hybrid Jeffreys+baseline-Gaussian configuration controls residual over-broad tails in the physical cold dark matter density $ω_{\mathrm{c}}$ while preserving the volume correction, and is our favoured approach. We compare the credible intervals derived using our methodology to those obtained using Halo Occupation Distribution (HOD)-informed priors and to confidence intervals derived using frequentist profile likelihood analyses, finding agreement in both central values and degeneracy directions in the $w_{0}$--$w_{a}$ plane. This demonstrates that, once projection effects are properly controlled, we can make robust inferences about the late-time cosmological expansion independent of the statistical framework adopted.

Bonici, M. [Waterloo U.; Perimeter Inst. Theor. Ph↗

A strategy for improved computational efficiency of the method of anchored distributions

This paper proposes a strategy for improving the computational efficiency of model inversion using the method of anchored distributions by “bundling” similar model parametrizations in the likelihood function. Inferring the likelihood function typically requires a large number of forward model simulations for each possible model parametrization; as a result, the process is quite expensive. To ease this prohibitive cost, we present an approximation for the likelihood function – called “bundling” – that relaxes the requirement for high quantities of forward model simulations. This approximation redefines the conditional statement of the likelihood function as the probability of sets of similar model parametrizations – “bundles” – replicating field measurements. To evaluate the effectiveness of these modifications, we compare the quality of predictions and computational cost of bundling relative to a baseline method of anchored distributions inversion of three-dimensional flow and transport model parameters. For our synthetic experiment, bundling achieved a 35% reduction in overall computational cost and had a limited negative on predicted probability distributions of the model parameters. Strategies for minimizing error in the bundling approximation, for enforcing similarity amongst the sets of model parametrizations, and for identifying convergence of the likelihood function are also presented.

Bayesian Inversion, SVD, computational efficiency↗

QMU as an Approach to Strengthening the Predictive Capabilities of Complex Models

Complex systems are made up of multiple interdependent parts, and the behavior of the entire system cannot always be directly inferred from the behavior of the individual parts. They are nonlinear and system responses are not necessarily additive. Examples of complex systems include energy, cyber and telecommunication infrastructures, human and animal social structures, and biological structures such as cells. To meet the goals of infrastructure development, maintenance, and protection for cyber-related complex systems, novel modeling and simulation technology is needed. Sandia has shown success using M&S in the nuclear weapons (NW) program. However, complex systems represent a significant challenge and relative departure from the classical M&S exercises, and many of the scientific and mathematical M&S processes must be re-envisioned. Specifically, in the NW program, requirements and acceptable margins for performance, resilience, and security are well-defined and given quantitatively from the start. The Quantification of Margins and Uncertainties (QMU) process helps to assess whether or not these safety, reliability and performance requirements have been met after a system has been developed. In this sense, QMU is used as a sort of check that requirements have been met once the development process is completed. In contrast, performance requirements and margins may not have been defined a priori for many complex systems, (i.e. the Internet, electrical distribution grids, etc.), particularly not in quantitative terms. This project addresses this fundamental difference by investigating the use of QMU at the start of the design process for complex systems. Three major tasks were completed. First, the characteristics of the cyber infrastructure problem were collected and considered in the context of QMU-based tools. Second, UQ methodologies for the quantification of model discrepancies were considered in the context of statistical models of cyber activity. Third, Bayesian methods for optimal testing in the QMU framework were developed. This completion of this project represent an increased understanding of how to apply and use the QMU process as a means for improving model predictions of the behavior of complex systems.

Gray, Genetha Anne↗

Bayesian network ensemble as a multivariate strategy to predict radiation pneumonitis risk

Purpose: Prediction of radiation pneumonitis (RP) has been shown to be challenging due to the involvement of a variety of factors including dose–volume metrics and radiosensitivity biomarkers. Some of these factors are highly correlated and might affect prediction results when combined. Bayesian network (BN) provides a probabilistic framework to represent variable dependencies in a directed acyclic graph. The aim of this study is to integrate the BN framework and a systems’ biology approach to detect possible interactions among RP risk factors and exploit these relationships to enhance both the understanding and prediction of RP. Methods: The authors studied 54 nonsmall-cell lung cancer patients who received curative 3D-conformal radiotherapy. Nineteen RP events were observed (common toxicity criteria for adverse events grade 2 or higher). Serum concentration of the following four candidate biomarkers were measured at baseline and midtreatment: alpha-2-macroglobulin, angiotensin converting enzyme (ACE), transforming growth factor, interleukin-6. Dose-volumetric and clinical parameters were also included as covariates. Feature selection was performed using a Markov blanket approach based on the Koller–Sahami filter. The Markov chain Monte Carlo technique estimated the posterior distribution of BN graphs built from the observed data of the selected variables and causality constraints. RP probability was estimated using a limited number of high posterior graphs (ensemble) and was averaged for the final RP estimate using Bayes’ rule. A resampling method based on bootstrapping was applied to model training and validation in order to control under- and overfit pitfalls. Results: RP prediction power of the BN ensemble approach reached its optimum at a size of 200. The optimized performance of the BN model recorded an area under the receiver operating characteristic curve (AUC) of 0.83, which was significantly higher than multivariate logistic regression (0.77), mean heart dose (0.69), and a pre-to-midtreatment change in ACE (0.66). When RP prediction was made only with pretreatment information, the AUC ranged from 0.76 to 0.81 depending on the ensemble size. Bootstrap validation of graph features in the ensemble quantified confidence of association between variables in the graphs where ten interactions were statistically significant. Conclusions: The presented BN methodology provides the flexibility to model hierarchical interactions between RP covariates, which is applied to probabilistic inference on RP. The authors’ preliminary results demonstrate that such framework combined with an ensemble method can possibly improve prediction of RP under real-life clinical circumstances such as missing data or treatment plan adaptation.

60 APPLIED LIFE SCIENCES↗

Uncertainty quantification in scientific machine learning: Methods, metrics, and comparisons

Neural networks (NNs) are currently changing the computational paradigm on how to combine data with mathematical laws in physics and engineering in a profound way, tackling challenging inverse and ill-posed problems not solvable with traditional methods. However, quantifying errors and uncertainties in NN-based inference is more complicated than in traditional methods. This is because in addition to aleatoric uncertainty associated with noisy data, there is also uncertainty due to limited data, but also due to NN hyperparameters, overparametrization, optimization and sampling errors as well as model misspecification. Although there are some recent works on uncertainty quantification (UQ) in NNs, there is no systematic investigation of suitable methods towards quantifying the total uncertainty effectively and efficiently even for function approximation, and there is even less work on solving partial differential equations and learning operator mappings between infinite-dimensional function spaces using NNs. In this work, we present a comprehensive framework that includes uncertainty modeling, new and existing solution methods, as well as evaluation metrics and post-hoc improvement approaches. Further, to demonstrate the applicability and reliability of our framework, we present an extensive comparative study in which various methods are tested on prototype problems, including problems with mixed input-output data, and stochastic problems in high dimensions. In the Appendix, we include a comprehensive description of all the UQ methods employed. Further, to help facilitate the deployment of UQ in Scientific Machine Learning research and practice, we present and develop in [1] an open-source Python library (github.com/Crunch-UQ4MI/neuraluq), termed NeuralUQ, that is accompanied by an educational tutorial and additional computational experiments.

11 physics-informed neural networks↗

The trailing edge is everywhere: tree rings reveal the transient risk of extinction hidden inside climate envelope forecasts

Given the importance of climate in shaping species’ geographic distributions, climate change poses an existential threat to biodiversity. Climate envelope modeling, the predominant approach used to quantify this threat, presumes that individuals in populations respond to climate variability and change according to species-level responses inferred from spatial occurrence data—such that individuals at the cool edge of a species’ distribution should benefit from warming (the “leading edge”), whereas individuals at the warm edge should suffer (the “trailing edge”). Using 1,558 tree-ring time series of an aridland pine (Pinus edulis) collected at 977 locations across the species’ distribution, we found that trees everywhere grow less in warmer-than-average and drier-than-average years. Ubiquitous negative temperature sensitivity indicates that individuals across the entire distribution should suffer with warming—the entire distribution is a trailing edge. Species-level responses to spatial climate variation are opposite in sign to individual-scale responses to time-varying climate for approximately half the species’ distribution with respect to temperature and the majority of the species’ distribution with respect to precipitation. These findings, added to evidence from the literature for scale-dependent climate responses in hundreds of species, suggest that correlative, equilibrium-based range forecasts may fail to accurately represent how individuals in populations will be impacted by changing climate. A scale-dependent view of the impact of climate change on biodiversity highlights the transient risk of extinction hidden inside climate envelope forecasts and the importance of evolution in rescuing species from extinction whenever local climate variability and change exceeds individual-scale climate tolerances.

54 ENVIRONMENTAL SCIENCES↗

Tree rings reveal the transient risk of extinction hidden inside climate envelope forecasts

Given the importance of climate in shaping species’ geographic distributions, climate change poses an existential threat to biodiversity. Climate envelope modeling, the predominant approach used to quantify this threat, presumes that individuals in populations respond to climate variability and change according to species-level responses inferred from spatial occurrence data—such that individuals at the cool edge of a species’ distribution should benefit from warming (the “leading edge”), whereas individuals at the warm edge should suffer (the “trailing edge”). Using 1,558 tree-ring time series of an aridland pine (Pinus edulis) collected at 977 locations across the species’ distribution, we found that trees everywhere grow less in warmer-than-average and drier-than-average years. Ubiquitous negative temperature sensitivity indicates that individuals across the entire distribution should suffer with warming—the entire distribution is a trailing edge. Species-level responses to spatial climate variation are opposite in sign to individual-scale responses to time-varying climate for approximately half the species’ distribution with respect to temperature and the majority of the species’ distribution with respect to precipitation. These findings, added to evidence from the literature for scale-dependent climate responses in hundreds of species, suggest that correlative, equilibrium-based range forecasts may fail to accurately represent how individuals in populations will be impacted by changing climate. A scale-dependent view of the impact of climate change on biodiversity highlights the transient risk of extinction hidden inside climate envelope forecasts and the importance of evolution in rescuing species from extinction whenever local climate variability and change exceeds individual-scale climate tolerances.

54 ENVIRONMENTAL SCIENCES↗

Inferring the source of evaporated waters using stable H and O isotopes

Stable isotope ratios of H and O are widely used to identify the source of water, e.g., in aquifers, river runoff, soils, plant xylem, and plant-based beverages. In situations where the sampled water is partially evaporated, its isotope values will have evolved along an evaporation line (EL) in δ 2 H/δ 18 O space, and back-correction along the EL to its intersection with a meteoric water line (MWL) has been used to estimate the source water’s isotope ratios. Here, we review the theory underlying isotopic estimation of source water for evaporated samples (iSW E ). We note potential for bias from a commonly used regression-based approach for EL slope estimation and suggest that a model-based approach may be preferable if assumptions of the regression approach are not valid. We then introduce a mathematical framework that eliminates the need to explicitly estimate the EL–MWL intersection, simplifying iSW E analysis and facilitating more rigorous uncertainty estimation. We apply this approach to data from the US EPA's 2007 National Lakes Assessment. We find that data for most lakes are consistent with a water source similar to annual runoff, estimated from monthly precipitation and evaporation within the lake basin. Strong evidence for both summer- and winter-biased sources exists, however, with winter bias pervasive in most snow-prone regions. Furthermore, the new analytical framework should improve the rigor of iSW E in ecohydrology and related sciences, and our initial results from US lakes suggest that previous interpretations of lakes as unbiased isotope integrators may only be valid in certain climate regimes.

54 ENVIRONMENTAL SCIENCES↗