Search NASA⌕ Search

SEARCH · Search NASA

Results for “Bayesian methods”

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 1,747 records · Page 97
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

BEYONDPLANCK XI. Bayesian CMB analysis with sample-based end-to-end error propagation

We present posterior sample-based cosmic microwave background (CMB) constraints from Planck LFI and WMAP observations as derived through global end-to-end Bayesian processing within the BeyondPlanck framework. We first use these samples to study correlations between CMB, foreground, and instrumental parameters, and we identify a particularly strong degeneracy between CMB temperature fluctuations and free-free emission on intermediate angular scales (400 ≲ ℓ ≲ 600), which is mitigated through model reduction, masking, and resampling. We compare our posterior-based CMB results with previous Planck products, and find generally good agreement, although with notably higher noise due to our exclusion of HFI data. We find a best-fit CMB dipole amplitude of 3362.7 ± 1.4 µK, in excellent agreement with previous Planck results. The quoted dipole uncertainty is derived directly from the sampled posterior distribution, and does not involve any ad hoc contributions for Planck instrumental systematic effects. Similarly, we find a temperature quadrupole amplitude of ${σ}_{2}^{TT}$ = 229 ± 97 µK 2 , which is in good agreement with previous results in terms of the amplitude, but the uncertainty is an order of magnitude larger than the naive diagonal Fisher uncertainty. Relatedly, we find lower evidence for a possible alignment between the quadrupole and octopole than previously reported due to a much larger scatter in the individual quadrupole coefficients, caused both by marginalizing over a more complete set of systematic effects, but also by our more conservative analysis mask required to mitigate the free-free degeneracy. For higher multipoles, we find that the angular temperature power spectrum is generally in good agreement with both Planck and WMAP. At the same time, we note that this is the first time the sample-based asymptotically exact Blackwell-Rao estimator has been successfully established for multipoles up to ℓ ≤ 600, and it now accounts for the majority of the cosmologically important information. Overall, this analysis demonstrates the unique capabilities of the Bayesian approach with respect to end-to-end systematic uncertainty propagation, and we believe it can and should play an important role in the analysis of future CMB experiments. Cosmological parameter constraints are presented in a companion paper.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Bayesian Characterization of Main-sequence Binaries in the Old Open Cluster NGC 188

The binary fractions of open and globular clusters yield powerful constraints on their dynamical state and evolutionary history. We apply publicly available Bayesian analysis tools to a UBV RIJHK {sub S} photometric catalog of the open cluster NGC 188 to detect and characterize photometric binaries along the cluster main sequence. This technique has the advantage of self-consistently handling photometric errors, missing data in various bandpasses, and star-by-star prior constraints on cluster membership. Simulations are used to verify uncertainties and quantify selection biases in our analysis, illustrating that among binaries with mass ratios >0.5, we recover the binary fraction to better than 7% in the mean, with no significant dependence on binary fraction and a mild dependence on assumed mass-ratio distribution. Using our photometric catalog, we recover the majority (65% ± 11%) of spectroscopically identified main-sequence binaries, including eight of the nine with spectroscopically measured mass ratios. Accounting for incompleteness and systematics, we derive a mass-ratio distribution that rises toward lower mass ratios (within our q > 0.5 analysis domain). We observe a raw binary fraction for solar-type main-sequence stars with mass ratios q > 0.5 of 42% ± 4%, independent of the assumed mass-ratio distribution to within its uncertainties, consistent with literature values for old open clusters but significantly higher than the field solar-type binary fraction. We confirm that the binaries identified by our method are more concentrated than single stars, in agreement with previous studies, and we demonstrate that the binary nature of those candidates that remain unidentified spectroscopically is strongly supported by photometry from Gaia DR2.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

Joint Planck and WMAP assessment of low CMB multipoles

The remarkable progress in cosmic microwave background (CMB) studies over past decade has led to the era of precision cosmology in striking agreement with the ΛCDM model. However, the lack of power in the CMB temperature anisotropies at large angular scales (low-ℓ), as has been confirmed by the recent Planck data also (up to 0ℓ=4), although statistically not very strong (less than 3σ), is still an open problem. One can avoid to seek an explanation for this problem by attributing the lack of power to cosmic variance or can look for explanations i.e., different inflationary potentials or initial conditions for inflation to begin with, non-trivial topology, ISW effect etc. Features in the primordial power spectrum (PPS) motivated by the early universe physics has been the most common solution to address this problem. In the present work we also follow this approach and consider a set of PPS which have features and constrain the parameters of those using WMAP 9 year and Planck data employing Markov-Chain Monte Carlo (MCMC) analysis. The prominent feature of all the models of PPS that we consider is an infra-red cut off which leads to suppression of power at large angular scales. We consider models of PPS with maximum three extra parameters and use Akaike information criterion (AIC) and Bayesian information criterion (BIC) of model selection to compare the models. For most models, we find good constraints for the cut off scale k{sub c}, however, for other parameters our constraints are not that good. We find that sharp cut off model gives best likelihood value for the WMAP 9 year data, but is as good as power law model according to AIC. For the joint WMAP 9 + Planck data set, Starobinsky model is slightly preferred by AIC which is also able to produce CMB power suppression up to 0ℓ≤3 to some extent. However, using BIC criteria, one finds model(s) with least number of parameters (power law model) are always preferred.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Risk-Informed Condition Evaluation of Solar-centered Energy Generation and Distribution Networks through Bayesian Learning and Inference

We develop a methodology based on Bayesian inference over Probabilistic Graphical Models (PGMs) to understand and quantify risk in solar-centered grids using targeted measurements and learned system behavior. Being non-prescriptive but, rather, able to infer system behavior and, ultimately, address risk queries from data, our machine learning-type paradigm is tailored for diverse topologies and threat scenarios often associated with distributed energy generation and photovoltaic distributed energy resources (PV-DERs) in particular. We describe algorithmic processes for: (i) learning the structure of PGMs that result from attack-prone PV-DER-proliferated distribution systems, (ii) quantifying cause-effect relationships, and (iii) evaluating risk queries based on diverse evidence. The contributions are illustrated on a residential grid subject to output impairment attacks on its PV-DER infrastructure.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Research Trends and Applications of PMUs

This work is a survey of current trends in applications of PMUs. PMUs have the potential to solve major problems in the areas of power system estimation, protection, and stability. A variety of methods are being used for these purposes, including statistical techniques, mathematical transformations, probability, and AI. The results produced by the techniques reviewed in this work are promising, but there is work to be performed in the context of implementation and standardization. As the smart grid initiative continues to advance, the number of intelligent devices monitoring the power grid continues to increase. PMUs are at the center of this initiative, and as a result, each year more PMUs are deployed across the grid. Since their introduction, myriad solutions based on PMU-technology have been suggested. The high sampling rates and synchronized measurements provided by PMUs are expected to drive significant advancements across multiple fields, such as the protection, estimation, and control of the power grid. This work offers a review of contemporary research trends and applications of PMU technology. Most solutions presented in this work were published in the last five years, and techniques showing potential for significant impact are highlighted in greater detail. Being a relatively new technology, there are several issues that must be addressed before PMU-based solutions can be successfully implemented. This survey found that key areas where improvements are needed include the establishment of PMU-observability, data processing algorithms, the handling of heterogeneous sampling rates, and the minimization of the investment in infrastructure for PMU communication. Solutions based on Bayesian estimation, as well as those having a distributed architectures, show great promise. The material presented in this document is tailored to both new researchers entering this field and experienced researchers wishing to become acquainted with emerging trends.

42 ENGINEERING↗
DOE OSTI2019

Likelihood non-Gaussianity in large-scale structure analyses

The standard present-day large-scale structure (LSS) analyses make a major assumption in their Bayesian parameter inference – that the likelihood has a Gaussian form. For summary statistics currently used in LSS, this assumption, even if the underlying density field is Gaussian, cannot be correct in detail. We investigate the impact of this assumption on two recent LSS analyses: the Beutler et al. power spectrum multipole ($P_ℓ$) analysis and the Sinha et al. group multiplicity function ($ζ$) analysis. Using non-parametric divergence estimators on mock catalogues originally constructed for covariance matrix estimation, we identify significant non-Gaussianity in both the $P_ℓ$ and $ζ$ likelihoods. We then use Gaussian mixture density estimation and independent component analysis on the same mocks to construct likelihood estimates that approximate the true likelihood better than the Gaussian pseudo-likelihood. Using these likelihood estimates, we accurately estimate the true posterior probability distribution of the Beutler et al. and Sinha et al. parameters. Likelihood non-Gaussianity shifts the fσ8 constraint by -0.44σ, but otherwise does not significantly impact the overall parameter constraints of Beutler et al. For the $ζ$ analysis, using the pseudo-likelihood significantly underestimates the uncertainties and biases the constraints of the Sinha et al. halo occupation parameters. For log $M_1$ and α, the posteriors are shifted by +0.43σ and -0.51σ and broadened by 42 per cent and 66 per cent, respectively. The divergence and likelihood estimation methods we present provide a straightforward framework for quantifying the impact of likelihood non-Gaussianity and deriving more accurate parameter constraints.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

A New Approach to Distributed Hypothesis Testing and Non-Bayesian Learning: Improved Learning Rate and Byzantine-Resilience

Here, we study a setting where a group of agents, each receiving partially informative private signals, seek to collaboratively learn the true underlying state of the world (from a finite set of hypotheses) that generates their joint observation profiles. To solve this problem, we propose a distributed learning rule that differs fundamentally from existing approaches, in that it does not employ any form of “belief-averaging”. Instead, agents update their beliefs based on a min-rule. Under standard assumptions on the observation model and the network structure, we establish that each agent learns the truth asymptotically almost surely. As our main contribution, we prove that with probability 1, each false hypothesis is ruled out by every agent exponentially fast, at a network-independent rate that is strictly larger than existing rates. We then develop a computationally-efficient variant of our learning rule that is provably resilient to agents who do not behave as expected (as represented by a Byzantine adversary model) and deliberately try to spread misinformation.

42 ENGINEERING↗
NASA NTRSDate not supplied

A Gaussian Process Enhancement to Linear Parameter Varying Models

Simulation and analysis for modern engineering systems now routinely requires the merging of multiple disciplines, physical-domains, time-scales, and data sets — all at ever increasing levels. These capabilities are especially needed in the domain of Advanced Air Mobility, where rapidly emerging vehicle designs are significantly more complex, while having to be both cost-effective and safe. To meet these engineering challenges, machine learning methods are an attractive option for merging models and data across multiple areas while providing uncertainty quantification and maintaining computational efficiency. This paper examines the use of Gaussian process machine learning to generalize and enhance the commonly used class of quasi-Linear Parameter Varying models for fast full-envelope simulation while also supporting control system design and analysis with model uncertainty. Gaussian process machine learning is selected because it: can fuse multiple data sets, enables an easy trade-off between data fitting and smoothing, provides model uncertainty quantification, scales well with increasing complexity, and does not generally require starting from a large training data set. To demonstrate the benefits of the approach, a robust stability analysis with Gaussian process uncertainty is shown for a NASA reference design of an electric quad-rotor air-taxi concept vehicle with motor parameter uncertainty.

Gaussian Process↗
DOE OSTI2015

Constraining Data Mining with Physical Models: Voltage- and Oxygen Pressure-Dependent Transport in Multiferroic Nanostructures

Development of new generation electronic devices requires understanding and controlling the electronic transport in ferroic, magnetic, and optical materials, which is hampered by two factors. First, the complications of working at the nanoscale, where interfaces, grain boundaries, defects, and so forth, dictate the macroscopic characteristics. Second, the convolution of the response signals stemming from the fact that several physical processes may be activated simultaneously. Here, we present a method of solving these challenges via a combination of atomic force microscopy and data mining analysis techniques. Rational selection of the latter allows application of physical constraints and enables direct interpretation of the statistically significant behaviors in the framework of the chosen physical model, thus distilling physical meaning out of raw data. We demonstrate our approach with an example of deconvolution of complex transport behavior in a bismuth ferrite–cobalt ferrite nanocomposite in ambient and ultrahigh vacuum environments. Measured signal is apportioned into four electronic transport patterns, showing different dependence on partial oxygen and water vapor pressure. These patterns are described in terms of Ohmic conductance and Schottky emission models in the light of surface electrochemistry. Finally and furthermore, deep data analysis allows extraction of local dopant concentrations and barrier heights empowering our understanding of the underlying dynamic mechanisms of resistive switching.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Assessing the performance of MagLIF with 3D MHD simulations

The Magnetized Liner Inertial Fusion (MagLIF) experimental platform at Sandia National Laboratories has realized a number of recent improvements in electrical current and laser preheat coupling but fusion yields are below expectations, based on modeling with clean (that is, simulations with no interfacial instabilities and no mix) two-dimensional (2D) magneto-hydrodynamics (MHD) simulations, by a factor of 5 or more. However, magnetized liner implosions are known to produce helical magneto-Rayleigh–Taylor (MRT) modes and complex stagnation structures that cannot be directly modeled in 2D. This paper presents the results of 3D HYDRA MHD simulations, including helical MRT, showing that degradation from these instabilities can readily reproduce experimental yields, but it is difficult to simultaneously match the yield along with inferred fuel temperature, pressure, and burn history (assessed with x-rays), which are larger in the experiments considered here. Additional analysis methods and future experiments are proposed to help address the discrepancies. The 3D simulations also show, without substantive improvements to stability, changing the applied Bz or preheat is unlikely to increase performance of MagLIF to the same degree as clean 2D simulations. Finally, the first 3D HYDRA simulations with the Hall term show the ab initio production of helical MRT and produce a clear change in stagnation morphology compared to pre-seeded simulations without Hall. In spite of the differences, the simulations still produce comparable fusion performance but with longer, lower power, neutron yield history.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2015

THE HUNT FOR EXOMOONS WITH KEPLER (HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS

We present a survey of 41 Kepler Objects of Interest (KOIs) for exomoons using Bayesian photodynamics, more than tripling the number of KOIs surveyed with this technique. We find no compelling evidence for exomoons although 13 KOIs yield spurious detections driven by instrumental artifacts, stellar activity, and/or perturbations from unseen bodies. Regarding the latter, we find seven KOIs exhibiting >5 σ evidence of transit timing variations, including the “mega-Earth” Kepler-10c, likely indicating an additional planet in that system. We exploit the moderately large sample of 57 unique KOIs surveyed to date to infer several useful statistics. For example, although there is a diverse range in sensitivities, we find that we are sensitive to Pluto–Charon mass-ratio systems for ≃40% of KOIs studied and Earth–Moon mass-ratios for 1 in 8 cases. In terms of absolute mass, our limits probe down to 1.7 Ganymede masses, with a sensitivity to Earth-mass moons for 1 in 3 cases studied and to the smallest moons capable of sustaining an Earth-like atmosphere (0.3 M{sub ⨁}) for 1 in 4. Despite the lack of positive detections to date, we caution against drawing conclusions yet, since our most interesting objects remain under analysis. Finally, we point out that had we searched for the photometric transit signals of exomoons alone, rather than using photodynamics, we estimate that 1 in 4 KOIs would have erroneously been concluded to harbor exomoons due to residual time correlated noise in the Kepler data, posing a serious problem for alternative methods.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS2010

Non-Targeted Effects and the Dose Response for Heavy Ion Tumorigenesis

BACKGROUND: There is no human epidemiology data available to estimate the heavy ion cancer risks experienced by astronauts in space. Studies of tumor induction in mice are a necessary step to estimate risks to astronauts. Previous experimental data can be better utilized to model dose response for heavy ion tumorigenesis and plan future low dose studies. DOSE RESPONSE MODELS: The Harderian Gland data of Alpen et al.[1-3] was re-analyzed [4] using non-linear least square regression. The data set measured the induction of Harderian gland tumors in mice by high-energy protons, helium, neon, iron, niobium and lanthanum with LET s ranging from 0.4 to 950 keV/micron. We were able to strengthen the individual ion models by combining data for all ions into a model that relates both radiation dose and LET for the ion to tumor prevalence. We compared models based on Targeted Effects (TE) to one motivated by Non-targeted Effects (NTE) that included a bystander term that increased tumor induction at low doses non-linearly. When comparing fitted models to the experimental data, we considered the adjusted R2, the Akaike Information Criteria (AIC), and the Bayesian Information Criteria (BIC) to test for Goodness of fit.In the adjusted R2test, the model with the highest R2values provides a better fit to the available data. In the AIC and BIC tests, the model with the smaller values of the summary value provides the better fit. The non-linear NTE models fit the combined data better than the TE models that are linear at low doses. We evaluated the differences in the relative biological effectiveness (RBE) and found the NTE model provides a higher RBE at low dose compared to the TE model. POWER ANALYSIS: The final NTE model estimates were used to simulate example data to consider the design of new experiments to detect NTE at low dose for validation. Power and sample sizes were calculated for a variety of radiation qualities including some not considered in the Harderian Gland data set and with different background tumor incidences. We considered different experimental designs with varying number of doses and varying low doses dependant on the LET of the radiation. The optimal design to detect a NTE for an individual ion had 4 doses equally spaced below a maximal dose where bending due to cell sterilization was < 2%. For example at 100 keV/micron we would irradiate at 0.03 Gy, 0.065 Gy, 0.13 Gy, and 0.26 Gy and require 850 mice including a control dose for a sensitivity to detect NTE with 80% power. Sample sizes could be improved by combining ions similar to the methods used with the Harderian Gland data.

Chappelli, Lori J.↗
DOE OSTI2023

Wave-function-based emulation for nucleon-nucleon scattering in momentum space

Emulators for low-energy nuclear physics can provide fast and accurate predictions of bound-state and scattering observables for applications that require repeated calculations with different parameters, such as Bayesian uncertainty quantification. In this paper, we extend a scattering emulator based on the Kohn variational principle (KVP) to momentum space (including coupled channels) with arbitrary boundary conditions, which enable the mitigation of spurious singularities known as Kohn anomalies. We test it on a modern chiral nucleon-nucleon (N N) interaction, including emulation of the coupled channels. We provide comparisons between a Lippmann-Schwinger equation emulator and our KVP momentum-space emulator for a representative set of neutron-proton (n p) scattering observables, and also introduce a quasi-spline-based approach for the KVP-based emulator. Furthermore, our findings show that while there are some trade-offs between accuracy and speed, all three emulators perform well. Self-contained Jupyter notebooks that generate the results and figures in this paper are publicly available.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2020

Predicting Biological Cleanliness: An Empirical Bayes Approach for Spacecraft Bioburden Accounting

To comply with the international planetary protection policy set forth by the Committee on Space Research and NASA Agency level requirements, spacecraft destined to biologically sensitive planetary bodies have to minimize terrestrial biological contamination. Analysis, testing and inspection are the standard forward verification activities that are used to demonstrate compliance with the biological contamination requirements. For testing of spacecraft surface areas, a swab or wipe sample is collected from surfaces prior to last access and subsequently processed in the lab using NASA Approved Planetary Protection Methods for Culture Based Assays. Raw data resulting from this assay is then statistically treated employing a mathematical paradigm stemming from the 1970’s Viking Lander Project to generate the bioburden density and total microbial bioburden present. This standard approach arbitrarily accounts for error and provides an upper conservative bound as it reports the maximum number of spores estimated to be present on flight hardware surfaces. A bioburden density estimate factors in the following variables: the observed bioburden count, representative volume processed, sampling efficiencies. Notably, to account for error in the approach, a 0 observed count is arbitrarily changed to a count of 1 for each hardware grouping. The data generated by spacecraft bioburden verification campaigns in the past have resulted in <80% of wipes and <90% of swabs containing a bioburden count of 0. As such, having a robust and well documented statistical approach for dealing with the probability of low incident rates is necessary to be able to estimate spacecraft bioburden. Being able to statistically describe the bioburden distribution and associated confidence level is a gamechanger for the development of bioburden allocations during mission design and will allow for tighter management of risk throughout spacecraft build. Thus, Empirical Bayes statistical approach was evaluated to estimate the microbial bioburden on spacecraft to mitigate the aforementioned mathematical concerns and provide a probabilistic bioburden distribution of the flight hardware surface. For application of this approach to performing bioburden calculations, a range of non-informative prior assumptions on hardware surfaces are explored for Bayesian analyses while informative priors using posterior distributions from prior assays are utilized for Empirical Bayes analyses. Several non-informative priors are currently under investigation to assess fitness including use of these priors to serve as a foundation to build off of NASA specification values or a basis of risk to account for unknowns during the integration and testing process. Informative priors under consideration are generated using sampled bioburden values from hardware originating within like processing environments (e.g. vendor cleaning process or similar assembly process), temporal spacecraft status events as a prediction for hardware cleanliness of future samples, and heritage system bioburden actuals to predict allocation for subsequent missions. Informative priors and probabilistic bioburden distributions are then validated using data sets from the Mars Exploration Rover, Mars Science Laboratory, and InSight missions. Using Empirical Bayes approach to generate a probabilistic bioburden distribution as demonstrated through mission use cases provides a valid approach for use in the end-to-end requirements verification process.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Determining H {sub 0} with Bayesian hyper-parameters

We re-analyse recent Cepheid data to estimate the Hubble parameter H {sub 0} by using Bayesian hyper-parameters (HPs). We consider the two data sets from Riess et al. 2011 and 2016 (labelled R11 and R16, with R11 containing less than half the data of R16) and include the available anchor distances (megamaser system NGC4258, detached eclipsing binary distances to LMC and M31, and MW Cepheids with parallaxes), use a weak metallicity prior and no period cut for Cepheids. We find that part of the R11 data is down-weighted by the HPs but that R16 is mostly consistent with expectations for a Gaussian distribution, meaning that there is no need to down-weight the R16 data set. For R16, we find a value of H {sub 0} = 73.75 ± 2.11 km s{sup −1} Mpc{sup −1} if we use HPs for all data points (including Cepheid stars, supernovae type Ia, and the available anchor distances), which is about 2.6 σ larger than the Planck 2015 value of H {sub 0} = 67.81 ± 0.92 km s{sup −1} Mpc{sup −1} and about 3.1 σ larger than the updated Planck 2016 value 66.93 ± 0.62 km s{sup −1} Mpc{sup −1}. If we perfom a standard χ{sup 2} analysis as in R16, we find H {sub 0} = 73.46 ± 1.40 (stat) km s{sup −1} Mpc{sup −1}. We test the effect of different assumptions, and find that the choice of anchor distances affects the final value significantly. If we exclude the Milky Way from the anchors, then the value of H {sub 0} decreases. We find however no evident reason to exclude the MW data. The HP method used here avoids subjective rejection criteria for outliers and offers a way to test datasets for unknown systematics.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Projective Integral Updates for High-Dimensional Variational Inference

Variational inference is an approximation framework for Bayesian inference that seeks to improve quantified uncertainty in predictions by optimizing a simplified distribution over parameters to stand in for the full posterior. Capturing model variations that remain consistent with training data enables more robust predictions by reducing parameter sensitivity. This work introduces a fixed-point optimization for variational inference that is applicable when every feasible log density can be expressed as a linear combination of functions from a given basis. In such cases, the optimizer becomes a fixed-point of projective integral updates. When the basis spans univariate quadratics in each parameter, the feasible distributions are Gaussian mean-fields and the projective integral updates yield quasi-Newton variational Bayes (QNVB). Other bases and updates are also possible. Since these updates require high-dimensional integration, this work begins by proposing an efficient quasirandom sequence of quadratures for mean-field distributions. Each iterate of the sequence contains two evaluation points that combine to correctly integrate all univariate quadratic functions and, if the mean-field factors are symmetric, all univariate cubics. More importantly, averaging results over short subsequences achieves periodic exactness on a much larger space of multivariate polynomials of quadratic total degree. The corresponding variational updates require four loss evaluations with standard (not second-order) backpropagation to eliminate error terms from over half of all multivariate quadratic basis functions. Furthermore, this integration technique is motivated by first proposing stochastic blocked mean-field quadratures, which may be useful in other contexts. A PyTorch implementation of QNVB allows for better control over model uncertainty during training than competing methods. Experiments demonstrate superior generalizability for multiple learning problems and architectures.

Gaussian mean-field↗
DOE OSTI2021

AT2017gfo: Bayesian inference and model selection of multicomponent kilonovae and constraints on the neutron star equation of state

The joint detection of the gravitational wave GW170817, of the short γ-ray burst GRB170817A and of the kilonova AT2017gfo, generated by the the binary neutron star (NS) merger observed on 2017 August 17, is a milestone in multimessenger astronomy and provides new constraints on the NS equation of state. We perform Bayesian inference and model selection on AT2017gfo using semi-analytical, multicomponents models that also account for non-spherical ejecta. Observational data favour anisotropic geometries to spherically symmetric profiles, with a log-Bayes’ factor of ~104, and favour multicomponent models against single-component ones. The best-fitting model is an anisotropic three-component composed of dynamical ejecta plus neutrino and viscous winds. Using the dynamical ejecta parameters inferred from the best-fitting model and numerical–relativity relations connecting the ejecta properties to the binary properties, we constrain the binary mass ratio to q < 1.54 and the reduced tidal parameter to $120\lt \tilde{\Lambda }\lt 1110$. Finally, we combine the predictions from AT2017gfo with those from GW170817, constraining the radius of a NS of 1.4 M ⊙ to 12.2 ± 0.5 km (1σ level). This prediction could be further strengthened by improving kilonova models with numerical-relativity information.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

BeyondPlanck: VII. Bayesian estimation of gain and absolute calibration for cosmic microwave background experiments

We present a Bayesian calibration algorithm for cosmic microwave background (CMB) observations as implemented within the global end-to-end BEYONDPLANCK framework and applied to the Planck Low Frequency Instrument (LFI) data. Following the most recent Planck analysis, we decomposed the full time-dependent gain into a sum of three nearly orthogonal components: one absolute calibration term, common to all detectors, one time-independent term that can vary between detectors, and one time-dependent component that was allowed to vary between one-hour pointing periods. Each term was then sampled conditionally on all other parameters in the global signal model through Gibbs sampling. The absolute calibration is sampled using only the orbital dipole as a reference source, while the two relative gain components were sampled using the full sky signal, including the orbital and Solar CMB dipoles, CMB fluctuations, and foreground contributions. We discuss various aspects of the data that influence gain estimation, including the dipole-polarization quadrupole degeneracy and processing masks. Comparing our solution to previous pipelines, we find good agreement in general, with relative deviations of -0.67% (-0.84%) for 30 GHz, 0.12% (-0.04%) for 44 GHz and -0.03% (-0.64%) for 70 GHz, compared to Planck PR4 and Planck 2018, respectively. We note that the BEYONDPLANCK calibration was performed globally, which results in better inter-frequency consistency than previous estimates. Additionally, WMAP observations were used actively in the BEYONDPLANCK analysis, which both breaks internal degeneracies in the Planck data set and results in an overall better agreement with WMAP. Finally, we used a Wiener filtering approach to smoothing the gain estimates. We show that this method avoids artifacts in the correlated noise maps as a result of oversmoothing the gain solution, which is difficult to avoid with methods like boxcar smoothing, as Wiener filtering by construction maintains a balance between data fidelity and prior knowledge. Although our presentation and algorithm are currently oriented toward LFI processing, the general procedure is fully generalizable to other experiments, as long as the Solar dipole signal is available to be used for calibration.

79 ASTRONOMY AND ASTROPHYSICS↗