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.

At least 1,063 records · Page 59

Spike-and-Slab Shrinkage Priors for Structurally Sparse Bayesian Neural Networks

Network complexity and computational efficiency have become increasingly significant aspects of deep learning. Sparse deep learning addresses these challenges by recovering a sparse representation of the underlying target function by reducing heavily overparameterized deep neural networks. Specifically, deep neural architectures compressed via structured sparsity (e.g., node sparsity) provide low-latency inference, higher data throughput, and reduced energy consumption. In this article, we explore two well-established shrinkage techniques, Lasso and Horseshoe, for model compression in Bayesian neural networks (BNNs). To this end, we propose structurally sparse BNNs, which systematically prune excessive nodes with the following: 1) spike-and-slab group Lasso (SS-GL) and 2) SS group Horseshoe (SS-GHS) priors, and develop computationally tractable variational inference, including continuous relaxation of Bernoulli variables. We establish the contraction rates of the variational posterior of our proposed models as a function of the network topology, layerwise node cardinalities, and bounds on the network weights. Furthermore, we empirically demonstrate the competitive performance of our models compared with the baseline models in prediction accuracy, model compression, and inference latency.

97 MATHEMATICS AND COMPUTING↗

A Variational Inference Approach to Inverse Problems with Gamma Hyperpriors

Hierarchical models with gamma hyperpriors provide a flexible, sparse-promoting framework to bridge L 1 and L 2 regularizations in Bayesian formulations to inverse problems. Despite the Bayesian motivation for these models, existing methodologies are limited to maximum a posteriori estimation. The potential to perform uncertainty quantification has not yet been realized. This paper introduces a variational iterative alternating scheme for hierarchical inverse problems with gamma hyperpriors. The proposed variational inference approach yields accurate reconstruction, provides meaningful uncertainty quantification, and is easy to implement. In addition, it lends itself naturally to conduct model selection for the choice of hyperparameters. Here, we illustrate the performance of our methodology in several computed examples, including a deconvolution problem and sparse identification of dynamical systems from time series data.

Bayesian shrinkage↗

Consequences of Asteroid Characterization on the State of Knowledge about Inferred Physical Properties and Impact Risk

Physical characteristics of Near-Earth Objects (NEOs) are essential inputs to planetary defense assessments. The size, density, and strength of an NEO are critical inputs to modeling behavior during atmospheric entry as well as assessing the risk of impact. Similarly, knowledge of the physical characteristics of an object are necessary to evaluate the probable result of a mitigation mission. Usually, these attributes cannot be directly measured, but increasingly sophisticated methods have been developed to infer physical properties from related measurements of asteroids, meteors, and/or meteorites. Fortuitously, some of these measurements have been obtained for enough NEOs to elucidate the distribution of values across the sampled population. However, the situation becomes more challenging when considering a specific asteroid, since it is unlikely that all the relevant measurements have been made for any given object. We have developed a Bayesian network that can combine available information about a particular NEO with knowledge of the larger population to infer probabilistic values and uncertainties for physical characteristics of interest. Distributions of asteroid population albedos, taxonomic classes, and macroporosities, along with meteorite density distributions and associations between taxonomic classes and meteorite classes, provide the default distributions for the network’s parameter nodes. The inference network links parameters for each virtual asteroid either deterministically or probabilistically as appropriate, and eliminates any unphysical combinations of parameters. Within the context of planetary defense, our Bayesian network can be used to constrain the ranges of likely impactor properties, which can subsequently reduce the uncertainty in modelling of atmospheric entry, mitigation efficacy, and impact risk assessment. When additional measurements become available for a specific object, the network incorporates those measurements to generate virtual asteroids with property distributions that are consistent with the measurements. We will use the 2023 PDC scenario to demonstrate how the inference network can be combined with plausible characterization measurements to refine the state of knowledge about likely combinations of physical parameters and the resulting impact risk.

risk assessment↗

Consequences of Asteroid Characterization on the State of Knowledge about Inferred Physical Properties and Impact Risk

Physical characteristics of Near-Earth Objects (NEOs) are essential inputs to planetary defense assessments. The size, density, and strength of an NEO are critical inputs to modeling behavior during atmospheric entry as well as assessing the risk of impact. Similarly, knowledge of the physical characteristics of an object are necessary to evaluate the probable result of a mitigation mission. Usually, these attributes cannot be directly measured, but increasingly sophisticated methods have been developed to infer physical properties from related measurements of asteroids, meteors, and/or meteorites. Fortuitously, some of these measurements have been obtained for enough NEOs to elucidate the distribution of values across the sampled population. However, the situation becomes more challenging when considering a specific asteroid, since it is unlikely that all the relevant measurements have been made for any given object. We have developed a Bayesian network that can combine available information about a particular NEO with knowledge of the larger population to infer probabilistic values and uncertainties for physical characteristics of interest. Distributions of asteroid population albedos, taxonomic classes, and macroporosities, along with meteorite density distributions and associations between taxonomic classes and meteorite classes, provide the default distributions for the network’s parameter nodes. The inference network links parameters for each virtual asteroid either deterministically or probabilistically as appropriate, and eliminates any unphysical combinations of parameters. Within the context of planetary defense, our Bayesian network can be used to constrain the ranges of likely impactor properties, which can subsequently reduce the uncertainty in modelling of atmospheric entry, mitigation efficacy, and impact risk assessment. When additional measurements become available for a specific object, the network incorporates those measurements to generate virtual asteroids with property distributions that are consistent with the measurements. We will use the 2023 PDC scenario to demonstrate how the inference network can be combined with plausible characterization measurements to refine the state of knowledge about likely combinations of physical parameters and the resulting impact risk.

risk assessment↗

Multiple model inference.

This paper compares three approaches for model selection: classical least squares methods, information theoretic criteria, and Bayesian approaches. Least squares methods are not model selection methods although one can select the model that yields the smallest sum-of-squared error function. Information theoretic approaches balance overfitting with model accuracy by incorporating terms that penalize more parameters with a log-likelihood term to reflect goodness of fit. Bayesian model selection involves calculating the posterior probability that each model is correct, given experimental data and prior probabilities that each model is correct. As part of this calculation, one often calibrates the parameters of each model and this is included in the Bayesian calculations. Our approach is demonstrated on a structural dynamics example with models for energy dissipation and peak force across a bolted joint. The three approaches are compared and the influence of the log-likelihood term in all approaches is discussed.

99 GENERAL AND MISCELLANEOUS↗

FlowPM: Distributed TensorFlow implementation of the FastPM cosmological N-body solver

Here, we present FlowPM, a Particle-Mesh (PM) cosmological N-body code implemented in Mesh-TensorFlow for GPU-accelerated, distributed, and differentiable simulations. We implement and validate the accuracy of a novel multi-grid scheme based on multiresolution pyramids to compute large-scale forces efficiently on distributed platforms. We explore the scaling of the simulation on large-scale supercomputers and compare it with corresponding Python based PM code, finding on an average 10x speed-up in terms of wallclock time. We also demonstrate how this novel tool can be used for efficiently solving large scale cosmological inference problems, in particular reconstruction of cosmological fields in a forward model Bayesian framework with hybrid PM and neural network forward model. We provide skeleton code for these examples and the entire code is publicly available at https://github.com/modichirag/flowpm[Formula presented].

79 ASTRONOMY AND ASTROPHYSICS↗

A scalable variational method for estimating the latent infection-rate field of an outbreak

In this paper, we explore whether the infection-rate of a disease can serve as a robust monitoring variable in epidemiological surveillance algorithms. The infection-rate is dependent on population mixing patterns that do not vary erratically day-to-day; in contrast, daily case-counts used in contemporary surveillance algorithms are corrupted by reporting errors. The technical challenge lies in estimating the latent infection-rate from case-counts. Here we devise a Bayesian method to estimate the infection-rate across multiple adjoining areal units, and then use it, via an anomaly detector, to discern a change in epidemiological dynamics. We extend an existing model for estimating the infection-rate in an areal unit by incorporating a Markov random field model, so that we may estimate infection-rates across multiple areal units, while preserving spatial correlations observed in the epidemiological dynamics. To carry out the high-dimensional Bayesian inverse problem, we develop an implementation of mean-field variational inference specific to the infection model and integrate it with the random field model to incorporate correlations across counties. The method is tested on estimating the COVID-19 infection-rates across all 33 counties in New Mexico using data from the summer of 2020, and then employing them to detect the arrival of the Fall 2020 COVID-19 wave. We perform the detection using a temporal algorithm that is applied county-by-county. We also show how the infection-rate field can be used to cluster counties with similar epidemiological dynamics.

60 APPLIED LIFE SCIENCES↗

Tuning a RANS k-e model for jet-in-crossflow simulations.

We develop a novel calibration approach to address the problem of predictive k e RANS simulations of jet-incrossflow. Our approach is based on the hypothesis that predictive k e parameters can be obtained by estimating them from a strongly vortical flow, specifically, flow over a square cylinder. In this study, we estimate three k e parameters, C%CE%BC, Ce2 and Ce1 by fitting 2D RANS simulations to experimental data. We use polynomial surrogates of 2D RANS for this purpose. We conduct an ensemble of 2D RANS runs using samples of (C%CE%BC;Ce2;Ce1) and regress Reynolds stresses to the samples using a simple polynomial. We then use this surrogate of the 2D RANS model to infer a joint distribution for the k e parameters by solving a Bayesian inverse problem, conditioned on the experimental data. The calibrated (C%CE%BC;Ce2;Ce1) distribution is used to seed an ensemble of 3D jet-in-crossflow simulations. We compare the ensemble's predictions of the flowfield, at two planes, to PIV measurements and estimate the predictive skill of the calibrated 3D RANS model. We also compare it against 3D RANS predictions using the nominal (uncalibrated) values of (C%CE%BC;Ce2;Ce1), and find that calibration delivers a significant improvement to the predictive skill of the 3D RANS model. We repeat the calibration using surrogate models based on kriging and find that the calibration, based on these more accurate models, is not much better that those obtained with simple polynomial surrogates. We discuss the reasons for this rather surprising outcome.

Lefantzi, Sophia↗

Improved Statistical Analysis for the Neutrinoless Double-Beta Decay Matrix Element of 136Xe

Neutrinoless double beta decay nuclear matrix element (M0ν) for 136Xe was recently analyzed using a statistical approach (Phys. Rev. C 107, 045501 (2023)). In the analysis, three initial shell model effective Hamiltonians were randomly altered, and their results for 23 measured observables were used to infer credibility for the M0ν nuclear matrix element (NME) based on a Bayesian Model Averaging approach. In that analysis, a reasonable Gamow-Teller quenching factor of 0.7 was assumed for each starting effective Hamiltonian. Given that the result of the statistical analysis was sensible to this choice, we are here improving that analysis by assuming that the Gamow-Teller quenching factor is also randomly chosen within reasonabe limits for all three starting Hamiltonians. The outcomes are slightly higher expectation values and uncertainties for the M0ν NME.

Astronomy & Astrophysics↗

ELG×LRG Distribution through Dark Matter Halo Dynamics

We investigate the clustering and halo occupation distribution (HOD) of DESI Y1 emission-line (ELGs) and luminous red (LRGs) galaxies at 0.8 < z < 1.1, including their cross-correlation (ELG×LRG), using the A BACUS S UMMIT suite and a new Halo Occupation Model (H OME ) for galaxy multitracers. This integrates intrahalo dynamics, halo exclusion, and quenching, bridging insights from hydrodynamical, HOD, abundance-matching, and semianalytic studies. Leveraging full phase-space information from the Uchuu N-body simulation, and sampling satellites from dark-matter particle positions via physically motivated prescriptions, Home reproduces the anisotropic clustering down to s = 200 h −1 kpc with unprecedented accuracy. Model parameters are inferred solely from two-point statistics using a two-level Bayesian framework, yielding high-fidelity ELG, LRG, and cross-reference catalogs. We find that satellite ELGs behave as incoherent flows within their parent halos, dominating the clustering below 4 h −1 Mpc. The HOD from the best-fit Home has the following properties: (i) 90.50% (85.91%) of ELGs (LRGs) are central galaxies without satellites, residing in halos of M vir ∼ 6.6 × 10 11 (1.2 × 10 13 ) h −1 M ⊙ ; (ii) the ELG×LRG cross-correlation is governed by central-central pairs and shaped by halo exclusion on 2–5 h −1 Mpc scales; (iii) 9.50% (14.09%) of ELGs (LRGs) are satellites, of which 1.09% (3.52%) inhabit halos with a central galaxy of the same species in a maximally conformal configuration, 7.02% (0.005%) orbit complementary hosts in a minimally conformal state, and 0.58% (10.57%) are orphans. The high sensitivity of Home precisely captures the dynamics of satellites in different host environments, opening a promising avenue for understanding systematics and the dynamical nature of dark matter, potentially distinguishing gravity models.

Favole, Ginevra [Universidad de La Laguna (Spain);↗

Past and present cosmic structure in the SDSS DR7 main sample

We present a chrono-cosmography project, aiming at the inference of the four dimensional formation history of the observed large scale structure from its origin to the present epoch. To do so, we perform a full-scale Bayesian analysis of the northern galactic cap of the Sloan Digital Sky Survey (SDSS) Data Release 7 main galaxy sample, relying on a fully probabilistic, physical model of the non-linearly evolved density field. Besides inferring initial conditions from observations, our methodology naturally and accurately reconstructs non-linear features at the present epoch, such as walls and filaments, corresponding to high-order correlation functions generated by late-time structure formation. Our inference framework self-consistently accounts for typical observational systematic and statistical uncertainties such as noise, survey geometry and selection effects. We further account for luminosity dependent galaxy biases and automatic noise calibration within a fully Bayesian approach. As a result, this analysis provides highly-detailed and accurate reconstructions of the present density field on scales larger than ∼ 3 Mpc/h, constrained by SDSS observations. This approach also leads to the first quantitative inference of plausible formation histories of the dynamic large scale structure underlying the observed galaxy distribution. The results described in this work constitute the first full Bayesian non-linear analysis of the cosmic large scale structure with the demonstrated capability of uncertainty quantification. Some of these results will be made publicly available along with this work. The level of detail of inferred results and the high degree of control on observational uncertainties pave the path towards high precision chrono-cosmography, the subject of simultaneously studying the dynamics and the morphology of the inhomogeneous Universe.

79 ASTRONOMY AND ASTROPHYSICS↗

Bayesian discovery of optimal reduced order models from mechanistic and experimental data: A case study of Pd penetration in TRISO fuels using BISON

TRistructural ISOtropic (TRISO) particles rely on a silicon carbide (SiC) layer as the primary structural material and barrier to metallic fission products (FPs) release. Accurate prediction of palladium (Pd) transport and penetration is therefore critical for qualifying TRISO fuels for advanced reactors. The empirical correlation for Pd penetration in BISON is derived from historical particle-fuel data, but cannot explain the large scatter in the experimental data that arises from varying experimental conditions. To aid fuel qualification, we previously developed a mechanistic reduced order model (ROM) using BISON that resolves these dependencies. Here, in this work we build on that mechanistic ROM and perform validation and quantify its uncertainty using Bayesian uncertainty quantification (UQ). calibration against a suite of in-pile and out-of-pile experiments spanning particle compositions, geometries, and operating conditions, and we benchmark it against the empirical correlation. Bayesian UQ identifies influential parameters, calibrates them to data, and yields predictive intervals. Results show that while the empirical correlation can be tuned to fit a single experiment type, it transfers poorly; the mechanistic ROM sustains accuracy with credible uncertainty across disparate conditions. This demonstrates a practical path—via Bayesian UQ applied to mechanistic ROMs—to leverage single-effect experiments for inferring in-reactor behavior and supporting TRISO fuel qualification.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Constraining the Milky Way Mass Profile with Phase-space Distribution of Satellite Galaxies

We estimate the Milky Way (MW) halo properties using satellite kinematic data including the latest measurements from Gaia DR2. With a simulation-based 6D phase-space distribution function (DF) of satellite kinematics, we can infer halo properties efficiently and without bias, and handle the selection function and measurement errors rigorously in the Bayesian framework. Applying our DF from the EAGLE simulation to 28 satellites, we obtain an MW halo mass of $M={1.23}_{-0.18}^{+0.21}\times {10}^{12}{M}_{\odot }$ and a concentration of $c={9.4}_{-2.1}^{+2.8}$ with the prior based on the M–c relation. The inferred mass profile is consistent with previous measurements but with better precision and reliability due to the improved methodology and data. Potential improvement is illustrated by combining satellite data and stellar rotation curves. Using our EAGLE DF and best-fit MW potential, we provide much more precise estimates of the kinematics for those satellites with uncertain measurements. Compared to the EAGLE DF, which matches the observed satellite kinematics very well, the DF from the semi-analytical model based on the dark-matter-only simulation Millennium II (SAM-MII) over-represents satellites with small radii and velocities. We attribute this difference to less disruption of satellites with small pericenter distances in the SAM-MII simulation. Finally, by varying the disruption rate of such satellites in this simulation, we estimate a ~5% scatter in the inferred MW halo mass among hydrodynamics-based simulations.

79 ASTRONOMY AND ASTROPHYSICS↗

Neural Posterior Estimation for Scalable and Accurate Inverse Parameter Inference in Li-Ion Batteries

Diagnosing the internal state of Li-ion batteries is critical for battery research, operation of real-world systems, and prognostic evaluation of remaining lifetime. By using physics-based models to perform probabilistic parameter estimation via Bayesian calibration, diagnostics can account for the uncertainty due to model fitness, data noise, and the observability of any given parameter. However, Bayesian calibration in Li-ion batteries using electrochemical data is computationally intensive even when using a fast surrogate in place of physics-based models, requiring many thousands of model evaluations. A fully amortized alternative is neural posterior estimation (NPE). NPE shifts the computational burden from the parameter estimation step to data generation and model training, reducing the parameter estimation time from minutes to milliseconds, enabling real-time applications. The present work shows that NPE can infer parameters equally or more accurately than Bayesian calibration, even if it leads to higher voltage reconstruction errors. We also demonstrate that the higher computational costs for data generation are tractable even in high-dimensional cases (ranging from 6 to 27 estimated parameters). The NPE method also offers several interpretability advantages over Bayesian calibration, such as local parameter sensitivity to specific regions of the voltage curve. The NPE method is demonstrated using an experimental fast charge dataset, with parameter estimates validated against measurements of loss of lithium inventory and loss of active material. The implementation is made available in a companion repository (https://github.com/NatLabRockies/BatFIT).

25 ENERGY STORAGE↗

Order priors for Bayesian network discovery with an application to malware phylogeny

Here, Bayesian networks have been used extensively to model and discover dependency relationships among sets of random variables. We learn Bayesian network structure with a combination of human knowledge about the partial ordering of variables and statistical inference of conditional dependencies from observed data. Our approach leverages complementary information from human knowledge and inference from observed data to produce networks that reflect human beliefs about the system as well as to fit the observed data. Applying prior beliefs about partial orderings of variables is an approach distinctly different from existing methods that incorporate prior beliefs about direct dependencies (or edges) in a Bayesian network. We provide an efficient implementation of the partial-order prior in a Bayesian structure discovery learning algorithm, as well as an edge prior, showing that both priors meet the local modularity requirement necessary for an efficient Bayesian discovery algorithm. In benchmark studies, the partial-order prior improves the accuracy of Bayesian network structure learning as well as the edge prior, even though order priors are more general. Our primary motivation is in characterizing the evolution of families of malware to aid cyber security analysts. For the problem of malware phylogeny discovery, we find that our algorithm, compared to existing malware phylogeny algorithms, more accurately discovers true dependencies that are missed by other algorithms.

97 MATHEMATICS AND COMPUTING↗

Evolution of homologous recombination rates across bacteria

Bacteria are nonsexual organisms but are capable of exchanging DNA at diverse degrees through homologous recombination. Intriguingly, the rates of recombination vary immensely across lineages where some species have been described as purely clonal and others as “quasi-sexual.” However, estimating recombination rates has proven a difficult endeavor and estimates often vary substantially across studies. It is unclear whether these variations reflect natural variations across populations or are due to differences in methodologies. Consequently, the impact of recombination on bacterial evolution has not been extensively evaluated and the evolution of recombination rate—as a trait—remains to be accurately described. Here, we developed an approach based on Approximate Bayesian Computation that integrates multiple signals of recombination to estimate recombination rates. We inferred the rate of recombination of 162 bacterial species and one archaeon and tested the robustness of our approach. Our results confirm that recombination rates vary drastically across bacteria; however, we found that recombination rate—as a trait—is conserved in several lineages but evolves rapidly in others. Although some traits are thought to be associated with recombination rate (e.g., GC-content), we found no clear association between genomic or phenotypic traits and recombination rate. Overall, our results provide an overview of recombination rate, its evolution, and its impact on bacterial evolution.

59 BASIC BIOLOGICAL SCIENCES↗

EXTRACTING THE SIZE OF THE COSMIC ELECTRON-POSITRON ANOMALY

We isolated the anomalous part of the cosmic electron-positron flux within a Bayesian likelihood analysis. Using 219 recent cosmic-ray spectral data points, we inferred the values of selected cosmic-ray propagation parameters. In the context of the propagation model coded in GalProp, we found a significant tension between the electron-positron related and the rest of the fluxes. Interpreting this tension as the presence of an anomalous component in the electron-positron related data, we calculated background predictions for PAMELA and Fermi-LAT based on the non-electron-positron related fluxes. We found a deviation between the data and the predicted background even when uncertainties, including systematics, were taken into account. We identified this deviation with the anomalous electron-positron contribution. We briefly compared this model-independent signal to some theoretical results predicting such an anomaly.

79 ASTRONOMY AND ASTROPHYSICS↗

Bayesian data assimilation for stochastic multiscale models of transport in porous media.

We investigate Bayesian techniques that can be used to reconstruct field variables from partial observations. In particular, we target fields that exhibit spatial structures with a large spectrum of lengthscales. Contemporary methods typically describe the field on a grid and estimate structures which can be resolved by it. In contrast, we address the reconstruction of grid-resolved structures as well as estimation of statistical summaries of subgrid structures, which are smaller than the grid resolution. We perform this in two different ways (a) via a physical (phenomenological), parameterized subgrid model that summarizes the impact of the unresolved scales at the coarse level and (b) via multiscale finite elements, where specially designed prolongation and restriction operators establish the interscale link between the same problem defined on a coarse and fine mesh. The estimation problem is posed as a Bayesian inverse problem. Dimensionality reduction is performed by projecting the field to be inferred on a suitable orthogonal basis set, viz. the Karhunen-Loeve expansion of a multiGaussian. We first demonstrate our techniques on the reconstruction of a binary medium consisting of a matrix with embedded inclusions, which are too small to be grid-resolved. The reconstruction is performed using an adaptive Markov chain Monte Carlo method. We find that the posterior distributions of the inferred parameters are approximately Gaussian. We exploit this finding to reconstruct a permeability field with long, but narrow embedded fractures (which are too fine to be grid-resolved) using scalable ensemble Kalman filters; this also allows us to address larger grids. Ensemble Kalman filtering is then used to estimate the values of hydraulic conductivity and specific yield in a model of the High Plains Aquifer in Kansas. Strong conditioning of the spatial structure of the parameters and the non-linear aspects of the water table aquifer create difficulty for the ensemble Kalman filter. We conclude with a demonstration of the use of multiscale stochastic finite elements to reconstruct permeability fields. This method, though computationally intensive, is general and can be used for multiscale inference in cases where a subgrid model cannot be constructed.

54 ENVIRONMENTAL SCIENCES↗