Search NASA⌕ Search

SEARCH · Search NASA

Results for “bayesian”

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 1,459 records · Page 81

Uncertainty quantification for Bayesian active learning in rupture life prediction of ferritic steels

Abstract Three probabilistic methodologies are developed for predicting the long-term creep rupture life of 9–12 wt%Cr ferritic-martensitic steels using their chemical and processing parameters. The framework developed in this research strives to simultaneously make efficient inference along with associated risk, i.e., the uncertainty of estimation. The study highlights the limitations of applying probabilistic machine learning to model creep life and provides suggestions as to how this might be alleviated to make an efficient and accurate model with the evaluation of epistemic uncertainty of each prediction. Based on extensive experimentation, Gaussian Process Regression yielded more accurate inference ( $$Pearson\;correlation\;coefficent> 0.95$$ P e a r s o n c o r r e l a t i o n c o e f f i c e n t > 0.95 for the holdout test set) in addition to meaningful uncertainty estimate (i.e., coverage ranges from 94 to 98% for the test set) as compared to quantile regression and natural gradient boosting algorithm. Furthermore, the possibility of an active learning framework to iteratively explore the material space intelligently was demonstrated by simulating the experimental data collection process. This framework can be subsequently deployed to improve model performance or to explore new alloy domains with minimal experimental effort.

20 FOSSIL-FUELED POWER PLANTS↗

Applying Bayesian inference and deterministic anisotropy to retrieve the molecular structure ∣Ψ(R)∣2 distribution from gas-phase diffraction experiments

Abstract Currently, our general approach to retrieving molecular structures from ultrafast gas-phase diffraction heavily relies on complex ab initio electronic or vibrational excited state simulations to make conclusive interpretations. Without such simulations, inverting this measurement for the structural probability distribution is typically intractable. This creates a so-called inverse problem. Here we address this inverse problem by developing a broadly applicable method that approximates the molecular frame structure ∣Ψ( R , t )∣ 2 distribution independent of these complex simulations. We retrieve the vibronic ground state ∣Ψ( R )∣ 2 for both simulated stretched NO 2 and measured N 2 O. From measured N 2 O, we observe 40 mÅ coordinate-space resolution from 3.75 Å −1 reciprocal space range and poor signal-to-noise, a 50X improvement over traditional Fourier transform methods. In simulated NO 2 diffraction experiments, typical to high signal-to-noise levels predict 100–1000X resolution improvements, down to 0.1 mÅ. By directly measuring the width of ∣Ψ( R )∣ 2 , we open ultrafast gas-phase diffraction capabilities to measurements beyond current analysis approaches. This method has the potential to effectively turn gas-phase ultrafast diffraction into a discovery-oriented technique to probe systems that are prohibitively difficult to simulate.

74 ATOMIC AND MOLECULAR PHYSICS↗

BEYONDPLANCK IV. On end-to-end simulations in CMB analysis — Bayesian versus frequentist statistics

End-to-end simulations play a key role in the analysis of any high-sensitivity cosmic microwave background (CMB) experiment, providing high-fidelity systematic error propagation capabilities that are unmatched by any other means. In this paper, we address an important issue regarding such simulations, namely, how to define the inputs in terms of sky model and instrument parameters. These may either be taken as a constrained realization derived from the data or as a random realization independent from the data. We refer to these as posterior and prior simulations, respectively. We show that the two options lead to significantly different correlation structures, as prior simulations (contrary to posterior simulations) effectively include cosmic variance, but they exclude realization-specific correlations from non-linear degeneracies. Consequently, they quantify fundamentally different types of uncertainties. We argue that as a result, they also have different and complementary scientific uses, even if this dichotomy is not absolute. In particular, posterior simulations are in general more convenient for parameter estimation studies, while prior simulations are generally more convenient for model testing. Before BEYONDPLANCK, most pipelines used a mix of constrained and random inputs and applied the same hybrid simulations for all applications, even though the statistical justification for this is not always evident. BEYONDPLANCK represents the first end-to-end CMB simulation framework that is able to generate both types of simulations and these new capabilities have brought this topic to the forefront. The BEYONDPLANCK posterior simulations and their uses are described extensively in a suite of companion papers. In this work, we consider one important applications of the corresponding prior simulations, namely, code validation. Specifically, we generated a set of one-year LFI 30 GHz prior simulations with known inputs and we used these to validate the core low-level BEYONDPLANCK algorithms dealing with gain estimation, correlated noise estimation, and mapmaking.

79 ASTRONOMY AND ASTROPHYSICS↗

Bayesian optimization of generalized data

Direct application of Bayes' theorem to generalized data yields a posterior probability distribution function (PDF) that is a product of a prior PDF of generalized data and a likelihood function, where generalized data consists of model parameters, measured data, and model defect data. The prior PDF of generalized data is defined by prior expectation values and a prior covariance matrix of generalized data that naturally includes covariance between any two components of generalized data. A set of constraints imposed on the posterior expectation values and covariances of generalized data via a given model is formally solved by the method of Lagrange multipliers. Posterior expectation values of the constraints and their covariance matrix are conventionally set to zero, leading to a likelihood function that is a Dirac delta function of the constraining equation. It is shown that setting constraints to values other than zero is analogous to introducing a model defect. Since posterior expectation values of any function of generalized data are integrals of that function over all generalized data weighted by the posterior PDF, all elements of generalized data may be viewed as nuisance parameters marginalized by this integration. One simple form of posterior PDF is obtained when the prior PDF and the likelihood function are normal PDFs. For linear models without a defect this PDF becomes equivalent to constrained least squares (CLS) method, that is, the χ2 minimization method.

Arbanas, Goran↗

Bayesian analysis of NmSuGra

In anticipation of data from the Large Hadron Collider (LHC) and the potential discovery of supersymmetry, we calculate the odds of the next-to-minimal version of the popular supergravity motivated model (NmSuGra) being discovered at the LHC to be 4:3 (57%). We also demonstrate that viable regions of the NmSuGra parameter space outside the LHC reach can be covered by upgraded versions of dark matter direct detection experiments, such as super-CDMS, at 95% confidence level. Due to the similarities of the models, we expect very similar results for the constrained minimal supersymmetric standard model (CMSSM).

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Towards a bayesian seismotectonic zoning for use in Probabilistic Seismic Hazard Assessment (PSHA)

The mathematical representation of seismic sources is an important part of probabilistic seismic hazard assessment. It reflects the association of the seismicity with the tectonically-active geological structures evidenced by seismotectonic studies. Given that most active faults are not characterized well enough, seismic sources are generally defined as areal zones, delimited with finite boundary polygons, within which the geological features of active tectonics and the seismicity are deemed homogeneous (e.g., focal depth, seismicity rate, and maximum magnitude). Besides the lack of data (e.g., narrow range of recorded magnitudes), the application of this representation generates different problems: 1) a large sensitivity of resulting hazard maps on the location of zone boundaries, while these boundaries are set by expert decision; 2) the zoning can not represent any variation in faulting mechanism; 3) the seismicity rates are distributed throughout the zones and we lose the location of the determinant information used for their calculation. We propose an exploratory study for an alternative procedure in area source modeling. First, different data (e.g., geomorphology, geology, fault orientations) will be combined by using automated spatial partitioning (investigation of both supervised and unsupervised methods) in order to obtain several information classes, which may be defined as areal source zones. Then, a given hypocenter belonging to a given ''zone'', from now on called seismicity model, will be expressed by a probability computed from the 2D (spatial) probability density function (pdf) for the active tectonic model used as an a priori and updated with specific data from seismicity catalogs (e.g., focal mechanism) or other new data sources (e.g., geomorphology, subsurface exploration). This hypocenter will thus be allowed to contribute to several models, with weights given by the value of the pdf for each model. The annual rate of occurrence, for a given model, will be calculated by the weighted average of the different hypocenter contributions contained in this model. Future applications will couple the seismicity models to Ground Motion Prediction Equations. In consequence, the results will provide the full spectrum of variability in the hazard and will highlight zones badly constrained and deserving to be more studied.

58 GEOSCIENCES↗

A Bayesian analysis of QCD sum rules

A new technique has recently been developed, in which the Maximum Entropy Method is used to analyze QCD sum rules. This approach has the virtue of being able to directly generate the spectral function of a given operator, without the need of making an assumption about its specific functional form. To investigate whether useful results can be extracted within this method, we have first studied the vector meson channel, where QCD sum rules are traditionally known to provide a valid description of the spectral function. Our results show a significant peak in the region of the experimentally observed {rho}-meson mass, which is in agreement with earlier QCD sum rules studies and suggests that the Maximum Entropy Method is a strong tool for analyzing QCD sum rules.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗