Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Generative Model”

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 757 records · Page 42
All sourcesNASA NTRSDOE OSTI
DOE OSTI2012

Spectroscopic detection of nitrogen concentrations in sagebrush

The ability to estimate foliar nitrogen (N) in semi-arid landscapes can yield information on nutritional status and improve our limited understanding of controls on canopy photosynthesis. We examined two spectroscopic methods for estimating sagebrush dried leaf and live shrub N content: first derivative reflectance (FDR) and continuum removal. Both methods used partial least squares (PLS) regression to select wavebands most significantly correlated with N concentrations in the samples. Sagebrush dried leaf spectra produced PLS models (R2 = 0.76–0.86) that could predict N concentrations within the dataset more accurately than PLS models generated from live shrub spectra (R2 = 0.41–0.63). Inclusion of wavelengths associated with leaf water in the FDR transformations appeared to improve regression results. Findings are encouraging and warrant further exploration into sagebrush reflectance spectra to characterize N concentrations.

58 GEOSCIENCES↗
DOE OSTI2015

POPULATION EFFECTS ON THE METALLICITY DISTRIBUTION FUNCTION DERIVED FROM THE RED GIANT BRANCH

We have tested the reliability of the red giant branch (RGB) as a metallicity indicator accounting for observational errors as well as the complexity of star formation histories and chemical evolution histories observed in various stellar systems. We generate model color–magnitude diagrams (CMDs) produced with a variety of evolutionary histories and compare the resultant metallicity estimates from the colors and magnitudes of RGB stars to the true input metallicities. We include realistic models for photometric errors and completeness in our synthetic CMDs. As expected, for simple simple stellar populations dominated by old stars, the RGB provides a very accurate estimate of the modular metallicity value for a population. An error in the age of a system targeted for this type of study may produce metallicity errors of a few tenths of a dex. The size of this metallicity error depends linearly on the age error, and we find this dependence to be stronger with more precise photometry. If the population has experienced any significant star formation within the last ∼6 Gyr, the metallicity estimates, [M/H], derived from the RGB may be in error by up to ∼0.5 dex. Perhaps the most important consideration for this technique is an accurate, independent estimate of the average age for the target stellar system, especially if it is probable that a significant fraction of the population formed less than ∼6 Gyr ago.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

Physical properties of Gliese 229B based on newly determined carbon and oxygen abundances of Gliese 229A

Recently Tsuji and Nakajima and Tsuji et al. have developed a method of molecular line spectroscopy of M dwarfs with which carbon and oxygen abundances are derived respectively from CO and H{sub 2}O lines in the K band. They applied this method to Gliese 229A(Gl 229A), the primary star of the brown dwarf companion, Gliese 229B(Gl 229B). The derived abundances of Gl 229A are logA{sub C}=−3.27±0.07 and logA{sub O}=−3.10±0.02, which are close to the classical values of the solar abundances of carbon and oxygen. We generate model spectra of Gl 229B for the metallicity of Gl 229A as well as for the classical solar metallicity (logA{sub C}=−3.40 and logA{sub O}=−3.08). We find that the differences in the resulting spectra are not so large for differences in metallicity of 0.1 dex or so, but we now discuss the spectrum of Gl 229B on the basis of reliable metallicity. From the literature, the lower limit to the age of Gl 229A is found to be 0.3 Gyr. From the kinematics of Gl 229A, we evaluate the upper limit to the age of Gl 229A to be 3.0 Gyr. The observed and model spectra are compared and the goodness of fit is obtained in the range of model parameters 750⩽T{sub eff}⩽1000 K and 4.5⩽logg⩽5.5. Among the candidates that satisfy the age constraint, the best combinations of model parameters are (T{sub eff},logg) = (800 K, 4.75) and (850 K, 5.0), while acceptable combinations are (750 K, 4.75), (850 K, 4.75) and (900 K, 5.0).

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2012

MAGNETIC FIELDS FROM QCD PHASE TRANSITIONS

We study the evolution of QCD phase transition-generated magnetic fields (MFs) in freely decaying MHD turbulence of the expanding universe. We consider an MF generation model that starts from basic non-perturbative QCD theory and predicts stochastic MFs with an amplitude of the order of 0.02 {mu}G and small magnetic helicity. We employ direct numerical simulations to model the MHD turbulence decay and identify two different regimes: a 'weakly helical' turbulence regime, when magnetic helicity increases during decay, and 'fully helical' turbulence, when maximal magnetic helicity is reached and an inverse cascade develops. The results of our analysis show that in the most optimistic scenario the magnetic correlation length in the comoving frame can reach 10 kpc with the amplitude of the effective MF being 0.007 nG. We demonstrate that the considered model of magnetogenesis can provide the seed MF for galaxies and clusters.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2015

TIDAL STREAM MORPHOLOGY AS AN INDICATOR OF DARK MATTER HALO GEOMETRY: THE CASE OF PALOMAR 5

This paper presents an example where the morphology of a single stellar stream can be used to rule out a specific galactic potential form without the need for velocity information. We investigate the globular cluster Palomar 5 (Pal 5), which is tidally disrupting into a cold, thin stream mapped over 22 deg on the sky with a typical width of 0.7 deg. We generate models of this stream by fixing Pal 5's present-day position, distance, and radial velocity via observations, while allowing its proper motion to vary. In a spherical dark matter halo we easily find models that fit the observed morphology. However, no plausible Pal 5 model could be found in the triaxial potential of Law and Majewski, which has been proposed to explain the properties of the Sagittarius stream. In this case, the long, thin, and curved morphology of the Pal 5 stream alone can be used to rule out such a potential configuration. Pal 5-like streams in this potential are either too straight, missing the curvature of the observations, or show an unusual morphology which we dub stream-fanning: a signature sensitive to the triaxiality of a potential. We conclude that the mere existence of other thin tidal streams must provide broad constraints on the orientation and shape of the dark matter halo they inhabit.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

Stability and coalescence of massive twin binaries

Massive stars are usually found in binaries, and binaries with periods less than 10 days may have a preference for near equal component masses (“twins”). In this paper we investigate the evolution of massive twin binaries all the way to contact and the possibility that these systems can be progenitors of double neutron star binaries. The small orbital separations of observed double neutron star binaries suggest that the progenitor systems underwent a common envelope phase at least once during their evolution. Bethe and Brown proposed that massive binary twins will undergo a common envelope evolution while both components are ascending the red giant branch (RGB) or asymptotic giant branch (AGB) simultaneously, also known as double-core evolution. Using models generated from the stellar evolution code EZ (evolve zero-age main sequence), we determine the range of mass ratios resulting in a contact binary with both components simultaneously ascending the RGB or AGB as a function of the difference in birth times, Δτ. We find that, even for a generous Δτ = 5 Myr, the minimum mass ratio q{sub min}=0.933 for an 8 M{sub ⊙} primary and increases for larger mass primaries. We use a smoothed particle hydrodynamics code, StarSmasher, to study specifically the evolution of q = 1 common envelope systems as a function of initial component mass, age, and orbital separation. We also consider a q = 0.997 system to test the effect of relaxing the constraint of strictly identical components. We find the dynamical stability limit, the largest orbital separation where the binary becomes dynamically unstable, as a function of the component mass and age. Finally, we calculate the efficiency of ejecting matter during the inspiral phase to extrapolate the properties of the remnant binary from our numerical results, assuming the common envelope is completely ejected. We find that for the nominal core masses, there is a minimum orbital separation for a given component mass such that the helium cores survive common envelope evolution in a tightly bound binary and are viable progenitors for double neutron stars.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

A NOVEL, FULLY AUTOMATED PIPELINE FOR PERIOD ESTIMATION IN THE EROS 2 DATA SET

We present a new method to discriminate periodic from nonperiodic irregularly sampled light curves. We introduce a periodic kernel and maximize a similarity measure derived from information theory to estimate the periods and a discriminator factor. We tested the method on a data set containing 100,000 synthetic periodic and nonperiodic light curves with various periods, amplitudes, and shapes generated using a multivariate generative model. We correctly identified periodic and nonperiodic light curves with a completeness of ∼90% and a precision of ∼95%, for light curves with a signal-to-noise ratio (S/N) larger than 0.5. We characterize the efficiency and reliability of the model using these synthetic light curves and apply the method on the EROS-2 data set. A crucial consideration is the speed at which the method can be executed. Using a hierarchical search and some simplification on the parameter search, we were able to analyze 32.8 million light curves in ∼18 hr on a cluster of GPGPUs. Using the sensitivity analysis on the synthetic data set, we infer that 0.42% of the sources in the LMC and 0.61% of the sources in the SMC show periodic behavior. The training set, catalogs, and source code are all available at http://timemachine.iic.harvard.edu.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2014

Supergravity inflation with broken shift symmetry and large tensor-to-scalar ratio

We propose a class of inflation models with potential V(φ) = αφ{sup n}exp(−β{sup m}φ{sup m}). We show that such kind of inflaton potentials can be realized in supergravity theory with a small shift symmetry breaking term in the K and quot;ahler potential. We find that the models with (m = 1,n = 1), (m = 1,n = 2), (m = 2,n = 1), and (m = 2,n = 3/2) can easily accommodate the Planck results. In particular, the tensor-to-scalar ratio is larger than 0.01 in the 1σ region. Thus, these models generate the typical large field inflation, and can be tested at the future Planck and QUBIC experiments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Field-level reconstruction from foreground-contaminated 21-cm maps

Current and upcoming 21-cm experiments will soon be able to map 21-cm spatial fluctuations in three dimensions for a wide range of redshifts. However, bright foreground contamination and the nature of radio interferometry create significant challenges, making it difficult to access rich cosmological information from the Fourier modes that lie within the “foreground wedge”. Here, in this work, we introduce two approaches aiming to reconstruct the full 21-cm density field, including the missing modes in the wedge: (a) a field-level inference under an effective field theory (EFT) framework; (b) a diffusion-based deep generative model trained on simulations. Under the EFT framework, we implement a fully differentiable forward model that maps the initial conditions of matter fluctuations to the observed, foreground-filtered 21-cm maps. This enables a gradient-based sampler to simultaneously sample the initial conditions and bias parameters, allowing a physically motivated mode reconstruction. Alternatively, we apply a variational diffusion model to perform 21-cm density reconstruction at the map level. Our model is trained on semi-numerical simulations over a wide range of astrophysical parameters. Our results from both approaches should provide improved cosmological constraints from the field level and also enable cross-correlation between experiments that have little or no overlapping modes.

cosmological perturbation theory↗
DOE OSTI2024

Robust anomaly detection for particle physics using multi-background representation learning

Abstract Anomaly, or out-of-distribution, detection is a promising tool for aiding discoveries of new particles or processes in particle physics. In this work, we identify and address two overlooked opportunities to improve anomaly detection (AD) for high-energy physics. First, rather than train a generative model on the single most dominant background process, we build detection algorithms using representation learning from multiple background types, thus taking advantage of more information to improve estimation of what is relevant for detection. Second, we generalize decorrelation to the multi-background setting, thus directly enforcing a more complete definition of robustness for AD. We demonstrate the benefit of the proposed robust multi-background AD algorithms on a high-dimensional dataset of particle decays at the Large Hadron Collider.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Discriminative versus generative approaches to simulation-based inference

Most of the fundamental, emergent, and phenomenological parameters of particle and nuclear physics are determined through parametric template fits. Simulations are used to populate histograms which are then matched to data. This approach is inherently lossy, since histograms are binned and low-dimensional. Deep learning has enabled unbinned and high-dimensional parameter estimation through neural likelihood(-ratio) estimation. We compare two approaches for neural simulation-based inference (NSBI): one based on discriminative learning (classification) and one based on generative modeling. These two approaches are directly evaluated on the same datasets, with a similar level of hyperparameter optimization in both cases. In addition to a Gaussian dataset, we study NSBI using a Higgs boson dataset from the FAIR Universe Challenge. We find that both the direct likelihood and likelihood ratio estimation are able to effectively extract parameters with reasonable uncertainties. For the numerical examples and within the set of hyperparameters studied, we found that the likelihood ratio method is more accurate and/or precise. Both methods have a significant spread from the network training and would require ensembling or other mitigation strategies in practice.

high energy physics↗
DOE OSTI2026

pop-cosmos : redshifts and physical properties of KiDS-1000 galaxies

ABSTRACT Principled Bayesian inference of galaxy properties has not previously been performed for wide-area weak-lensing surveys with millions of sources. We address this gap by applying the pop-cosmos generative model to perform spectral energy distribution (SED) fitting for 4 million KiDS (Kilo-Degree Survey)-1000 galaxies. Calibrated on deep COSMOS2020 photometric data, pop-cosmos specifies a physically motivated prior over the galaxy population up to $z \simeq 6$ in stellar population synthesis (SPS) parameter space. Using the Speculator SPS emulator with GPU (graphics processing unit)-accelerated Markov Chain Monte Carlo sampling, we perform full posterior inference at 8.2 GPU seconds per galaxy, obtaining joint constraints on galaxy redshifts and physical properties. We validate photometric redshifts against $\sim \!185\,\!000$ KiDS galaxies cross-matched to Dark Energy Spectroscopic Instrument Data Release 1 spectroscopic samples, achieving low bias ($2\times 10^{-3}$), scatter ($\sigma _{\mathrm{MAD}}=0.03$), and outlier fraction (3.2 per cent) for the Bright Galaxy Survey, with comparable performance (bias $3\times 10^{-2}$, $\sigma _{\mathrm{MAD}}=0.05$, 1.0 per cent outliers) for luminous red galaxies (LRGs). Within the LRG sample, we identify massive, dusty, star-forming contaminants at $z \simeq 0.4$ satisfying standard colour selections for quenched populations. We infer trends in stellar mass, star formation, metallicity, and dust across five tomographic redshift bins consistent with established scaling relations. Using specific star formation rate constraints, we identify $\sim$7 per cent of KiDS-1000 galaxies as quenched, versus 37 per cent implied by conservative colour cuts. This enables the construction of weak-lensing samples defined by physical properties while mitigating intrinsic alignment systematics and preserving statistical power. Our analysis validates pop-cosmos out of sample, establishing it as a scalable approach for galaxy evolution and cosmological analyses with photometric surveys.

Halder, Anik [Institute of Astronomy and Kavli Ins↗
DOE OSTI2019

Deblending galaxy superpositions with branched generative adversarial networks

Abstract Near-future large galaxy surveys will encounter blended galaxy images at a fraction of up to 50 per cent in the densest regions of the Universe. Current deblending techniques may segment the foreground galaxy while leaving missing pixel intensities in the background galaxy flux. The problem is compounded by the diffuse nature of galaxies in their outer regions, making segmentation significantly more difficult than in traditional object segmentation applications. We propose a novel branched generative adversarial network to deblend overlapping galaxies, where the two branches produce images of the two deblended galaxies. We show that generative models are a powerful engine for deblending given their innate ability to infill missing pixel values occluded by the superposition. We maintain high peak signal-to-noise ratio and structural similarity scores with respect to ground truth images upon deblending. Our model also predicts near-instantaneously, making it a natural choice for the immense quantities of data soon to be created by large surveys such as Large Synoptic Survey Telescope, Euclid, and Wide-Field Infrared Survey Telescope.

Reiman, David M.↗
DOE OSTI2011

High regularity of Z-DNA revealed by ultra high-resolution crystal structure at 0.55;#8201;Å

The crystal structure of a Z-DNA hexamer duplex d(CGCGCG){sub 2} determined at ultra high resolution of 0.55 {angstrom} and refined without restraints, displays a high degree of regularity and rigidity in its stereochemistry, in contrast to the more flexible B-DNA duplexes. The estimations of standard uncertainties of all individually refined parameters, obtained by full-matrix least-squares optimization, are comparable with values that are typical for small-molecule crystallography. The Z-DNA model generated with ultra high-resolution diffraction data can be used to revise the stereochemical restraints applied in lower resolution refinements. Detailed comparisons of the stereochemical library values with the present accurate Z-DNA parameters, shows in general a good agreement, but also reveals significant discrepancies in the description of guanine-sugar valence angles and in the geometry of the phosphate groups.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Probing nuclear effects with neutrino-induced charged-current neutral pion production

We study neutrino-induced charged-current (CC) $\pi^0$ production on carbon nuclei using events with fully imaged final-state proton-$\pi^0$ systems. Novel use of final-state correlations based on transverse kinematic imbalance enable the first measurements of the struck nucleon's Fermi motion, of the intranuclear momentum transfer (IMT) dynamics, and of the final-state hadronic momentum configuration in neutrino pion production. Event distributions are presented for i) the momenta of neutrino-struck neutrons below the Fermi surface, ii) the direction of missing transverse momentum characterizing the strength of IMT, and iii) proton-pion momentum imbalance with respect to the lepton scattering plane. The observed Fermi motion and IMT strength are compared to the previous MINERvA measurement of neutrino CC quasielastic-like production. The measured shapes and absolute rates of these distributions, as well as the cross-section asymmetries show tensions with predictions from current neutrino generator models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Flow-based likelihoods for non-Gaussian inference

We investigate the use of data-driven likelihoods to bypass a key assumption made in many scientific analyses, which is that the true likelihood of the data is Gaussian. In particular, we suggest using the optimization targets of flow-based generative models, a class of models that can capture complex distributions by transforming a simple base distribution through layers of nonlinearities. We call these flow-based likelihoods (FBL). We analyze the accuracy and precision of the reconstructed likelihoods on mock Gaussian data, and show that simply gauging the quality of samples drawn from the trained model is not a sufficient indicator that the true likelihood has been learned. We nevertheless demonstrate that the likelihood can be reconstructed to a precision equal to that of sampling error due to a finite sample size. We then apply FBLs to mock weak lensing convergence power spectra, a cosmological observable that is significantly non-Gaussian (NG). We find that the FBL captures the NG signatures in the data extremely well, while other commonly used data-driven likelihoods, such as Gaussian mixture models and independent component analysis, fail to do so. This suggests that works that have found small posterior shifts in NG data with data-driven likelihoods such as these could be underestimating the impact of non-Gaussianity in parameter constraints. By introducing a suite of tests that can capture different levels of NG in the data, we show that the success or failure of traditional data-driven likelihoods can be tied back to the structure of the NG in the data. Here, unlike other methods, the flexibility of the FBL makes it successful at tackling different types of NG simultaneously. Because of this, and consequently their likely applicability across datasets and domains, we encourage their use for inference when sufficient mock data are available for training.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Nonequilibrium effects on stability of hybrid stars with first-order phase transitions

The stability of hybrid stars with first-order phase transitions as determined by calculating fundamental radial oscillation modes is known to differ from the predictions of the widely used Bardeen-Thorne-Meltzer criterion. We consider the effects of out-of-chemical-equilibrium physics on the radial modes and hence stability of these objects. For a barotropic equation of state, this is done by allowing the adiabatic sound speed to differ from the equilibrium sound speed. We show that doing so extends the stable branches of stellar models, allowing stars with rapid phase transitions to support stable higher-order stellar multiplets similarly to stars with multiple slow phase transitions. We also derive a new junction condition to impose on the oscillation modes at the phase transition. Termed the reactive condition, it is physically motivated, consistent with the generalized junction conditions between two phases, and has the common rapid and slow conditions as limiting cases. Unlike the two common cases, it can only be applied to nonbarotropic stars. Here, we apply this junction condition to hybrid stellar models generated using a two-phase equation of state consisting of nuclear matter with unpaired quark matter at high densities joined by a first-order phase transition and show that, like in the slow limiting case, stars that are classically unstable are stabilized by a finite chemical reaction speed.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2010

Fourth generation effect on {Lambda}{sub b} decays

The rare decays of the {Lambda}{sub b} baryon governed by the quark level transitions b{yields}s are investigated in the fourth quark generation model popularly known as SM4. Recently it has been shown that SM4, which is a very simple extension of the standard model, can successfully explain several anomalies observed in the CP violation parameters of B and B{sub s} mesons. We find that in this model due to the additional contributions coming from the heavy t{sup '} quark in the loop, the branching ratios and other observables in rare {Lambda}{sub b} decays deviate significantly from their standard model values. Some of these modes are within the reach of the LHCb experiment and search for such channels is strongly argued.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗