Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “inference accelerators”

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 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Unraveling emission line galaxy conformity at z ∼ 1 with DESI early data

Emission line galaxies (ELGs) are now the preeminent tracers of large-scale structure at z > 0.8 due to their high density and strong emission lines, which enable accurate redshift measurements. However, relatively little is known about ELG evolution and the ELG–halo connection, exposing us to potential modelling systematics in cosmology inference using these sources. In this paper, we use a variety of observations and simulated galaxy models to propose a physical picture of ELGs and improve ELG–halo connection modelling in a halo occupation distribution framework. We investigate Dark Energy Spectroscopic Instrument (DESI)-selected ELGs in COSMOS data, and infer that ELGs are rapidly star-forming galaxies with a large fraction exhibiting disturbed morphology, implying that many of them are likely to be merger-driven starbursts. We further postulate that the tidal interactions from mergers lead to correlated star formation in central–satellite ELG pairs, a phenomenon dubbed ‘conformity’. We argue for the need to include conformity in the ELG–halo connection using galaxy models such as IllustrisTNG, and by combining observations such as the DESI ELG autocorrelation, ELG cross-correlation with luminous red galaxies, and ELG–cluster cross-correlation. We also explore the origin of conformity using the UniverseMachine model and elucidate the difference between conformity and the well-known galaxy assembly bias effect.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

A continuous calibration of the ATLAS flavour-tagging classifiers via optimal transportation maps

A calibration of the ATLAS flavour-tagging algorithms using a new calibration procedure based on optimal transportation maps is presented. Simultaneous, continuous corrections to the b-jet, c-jet, and light-flavour jet classification probabilities from jet-tagging algorithms in simulation are derived for b-jets using $t\bar{t} \rightarrow e\mu \nu \nu bb$ data. After application of the derived calibration maps, closure between simulation and observation is achieved for jet flavour observables used in ATLAS analyses of Large Hadron Collider (LHC) Run 2 proton-proton collision data. This continuous calibration opens up new possibilities for the future use of jet flavour information in LHC analyses and also serves as a guide for deriving high-dimensional corrections to simulation via transportation maps, an important development for a broad range of inference tasks.

Aad, G. [Aix-Marseille Université] (ORCID:00000002↗
DOE OSTI2026

Bayesian Exploration and Surrogate Emulation of Nonlinear Beam-Response Geometry in the LBNF Beamline

Next-generation long-baseline neutrino experiments aim to achieve multi-MW proton beam power while reducing accelerator-induced systematic uncertainties. At Fermilab, the LBNF beamline is designed for 1.2 MW operation with PIP-II and is upgradeable to 2.4 MW. DUNE will probe the three-flavor neutrino paradigm and search for CP violation, requiring precise neutrino-flux normalization and improved control of accelerator-related uncertainties. Within the LBNF beamline, the System for On-Axis Neutrino Detection (SAND) will constrain flux uncertainties using precision near-detector measurements, while the Muon Monitor System (MuMS) will provide beamline diagnostics sensitive to the proton beam, target, and horn configuration. However, the pion phase space relevant for DUNE depends simultaneously on many correlated parameters, including beam centroid, beam width, horn current and alignment, target position, optics shifts, and radiation-induced changes. Consequently, MuMS observables exhibit nonlinear and coupled responses that are difficult to characterize using traditional one-parameter scans. To address this challenge, we are developing a Bayesian Exploration framework coupled to physics-informed surrogate emulators trained on Geant4 beamline simulations. Gaussian-process emulators provide both fast predictions and uncertainty estimates, enabling adaptive selection of new simulation points in beam-parameter space. As an initial demonstration, we construct surrogate emulators for MuMS response observables using a verified simulation campaign spanning proton-beam steering conditions. The emulators reproduce the simulated dependence of MuMS centroid and gradient observables while providing predictive uncertainties, and serve as the foundation for future multidimensional exploration including beam width, horn current, and additional beamline parameters. This work establishes a framework for uncertainty-aware beam monitoring, adaptive simulation campaigns, and rapid beam-response inference for future DUNE operations.

Ganguly, Sudeshna [Fermilab] (ORCID:00000003163482↗
DOE OSTI2025

Deep inference of simulated strong lenses in ground-based surveys

The large number of strong lenses discoverable in future astronomical surveys will likely enhance the value of strong gravitational lensing as a cosmic probe of dark energy and dark matter. However, leveraging the increased statistical power of such large samples will require further development of automated lens modeling techniques. We show that deep learning and simulation-based inference (SBI) methods produce informative and reliable estimates of parameter posteriors for strong lensing systems in ground-based surveys. We present the examination and comparison of two approaches to lens parameter estimation for strong galaxy-galaxy lenses — Neural Posterior Estimation (NPE) and Bayesian Neural Networks (BNNs). We perform inference on 1-, 5-, and 12-parameter lens models for ground-based imaging data that mimics the Dark Energy Survey (DES). We find that NPE outperforms BNNs, producing posterior distributions that are more accurate, precise, and well-calibrated for most parameters. For the 12-parameter NPE model, the calibration is consistently within <10% of optimal calibration for all parameters, while the BNN is rarely within 20% of optimal calibration for any of the parameters. Similarly, residuals for most of the parameters are smaller (by up to an order of magnitude) with the NPE model than the BNN model. This work takes important steps in the systematic comparison of methods for different levels of model complexity.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Can local white dwarfs constrain dark matter interactions?

We investigate whether nearby white dwarfs (WDs) can constrain dark matter (DM) interactions with ordinary matter. As experimental sensitivity improves, driven by the Gaia mission, the sample volume of nearby WDs has been increasing over recent years. We carefully select a sample of ten cold, isolated, non-magnetic WDs within 13 pc of the Sun. We model their carbon-oxygen interior using a finite temperature relativistic equation of state and model atmospheres to infer their core temperatures. This enables us to perform a thorough estimation of the DM capture rate and evaporation mass using actual astrophysical observations. Given the low local DM density, we focus on DM that annihilates into long-lived mediators, which escape the WD and later decay into photons. While Fermi-LAT data shows no significant gamma-ray excess, future telescopes, CTA North & South, LHAASO, SWGO, could probe DM-nucleon cross sections down to ∼ 10 -41 cm 2 for DM masses above the TeV scale. Our results are competitive with current direct detection bounds (e.g., LZ) in the multi-TeV regime. This work underscores the importance of systematic WD studies in the broader landscape of DM detection and demonstrates the synergy between astrophysical and terrestrial searches in exploring DM interactions.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

SAGAbg. III. Environmental Stellar Mass Functions, Self-quenching, and the Stellar-to-halo Mass Relation in the Dwarf Galaxy Regime

Recent efforts have extended our view of the number and properties of satellite galaxies beyond the Local Group firmly down to M ⋆ ∼ 10 6 M ⊙ . A similarly complete view of the field dwarf population has lagged behind. Using the background galaxy sample from the Satellites Around Galactic Analogs (SAGA) survey at z < 0.05, we take inventory of the dwarf population down to M ⋆ ∼ 5 × 10 6 M ⊙ using three metrics: the stellar mass function (SMF) as a function of environment, the stellar-to-halo mass relation (SHMR) of dwarf galaxies inferred via abundance matching, and the quenched fraction of highly isolated dwarfs. We find that the low-mass SMF shape shows minimal environmental dependence, with the field dwarf SMF described by a low-mass power-law index of α 1 = −1.44 ± 0.09 down to M ⋆ ∼ 5 × 10 6 M ⊙ , and that the quenched fraction of isolated dwarfs drops monotonically to f q ∼ 10 −3 at M ⋆ ∼ 10 8.5 M ⊙ . Though slightly steeper than estimates from H I kinematic measures, our inferred SHMR agrees with literature measurements of satellite systems, consistent with minimal environmental dependence of the SHMR in the probed mass range. Finally, although most contemporary cosmological simulations against which we compare accurately predict the SAGAbg-SMF SHMR, we find that big-box cosmological simulations largely overpredict isolated galaxy quenched fractions via a turnaround in f q (M ⋆ ) at 10 8 ≲ M ⋆ /M ⊙ ≲ 10 9 , underscoring the complexities in disentangling the drivers of galaxy formation and the need for systematic multidimensional observations of the dwarf population across environments.

Kado-Fong, Erin [Yale Univ., New Haven, CT (United↗
DOE OSTI2026

Gaussian processes for inferring parton distributions

The extraction of parton distribution functions (PDFs) from experimental or lattice QCD data is an ill-posed inverse problem, where regularization strongly impacts both systematic uncertainties and the reliability of the results. We study a framework based on Gaussian Process Regression (GPR) to reconstruct PDFs from lattice QCD matrix elements. Within a Bayesian framework, Gaussian processes serve as flexible priors that encode uncertainties, correlations, and constraints without imposing rigid functional forms. We investigate a wide range of kernel choices, mean functions, and hyperparameter treatments. We quantify information gained from the data using the Kullback-Leibler divergence. Synthetic data tests demonstrate the consistency and robustness of the method. Our study establishes GPR as a systematic and non-parametric approach to PDF reconstruction, offering controlled uncertainty estimates and reduced model bias in lattice QCD analyses.

hadronic spectroscopy↗
DOE OSTI2021

Multilevel Hierarchical Decomposition of Finite Element White Noise with Application to Multilevel Markov Chain Monte Carlo

In this work we develop a new hierarchical multilevel approach to generate Gaussian random field realizations in an algorithmically scalable manner that is well suited to incorporating into multilevel Markov chain Monte Carlo (MCMC) algorithms. This approach builds off of other partial differential equation (PDE) approaches for generating Gaussian random field realizations; in particular, a single field realization may be formed by solving a reaction-diffusion PDE with a spatial white noise source function as the right-hand side. While these approaches have been explored to accelerate forward uncertainty quantification tasks, e.g., multilevel Monte Carlo, the previous constructions are not directly applicable to multilevel MCMC frameworks which build fine-scale random fields in a hierarchical fashion from coarse-scale random fields. Our new hierarchical multilevel method relies on a hierarchical decomposition of the white noise source function in $L^2$ which allows us to form Gaussian random field realizations across multiple levels of discretization in a way that fits into multilevel MCMC algorithmic frameworks. After presenting our main theoretical results and numerical scaling results to showcase the utility of this new hierarchical PDE method for generating Gaussian random field realizations, this method is tested on a four-level MCMC algorithm to explore its feasibility.

algebraic multigrid↗
DOE OSTI2025

Study of the connected four-point correlation function of galaxies from the DESI Data Release 1 luminous red galaxy sample

We present a measurement of the non-Gaussian four-point correlation function (4PCF) from the DESI DR1 luminous red galaxy (LRG) sample. For the gravitationally induced parity-even 4PCF, we detect a signal with a significance of 14.7⁢𝜎 using our fiducial setup. We assess the robustness of this detection through a series of validation tests, including auto and cross-correlation analyses, sky partitioning across multiple patch combinations, and variations in radial scale cuts. Due to the low completeness of the sample, we find that differences in fiber assignment implementation schemes can significantly impact estimation of the covariance and introduce biases in the data vector. After correcting for these effects, all tests yield consistent results. This is one of the first measurements of the connected 4PCF on the DESI LRG sample; the good agreement between the simulation and the data implies that the amplitude of the density fluctuation inferred from the connected 4PCF is consistent with the Planck Λ⁢ CDM cosmology. The methodology and diagnostic framework established in this work provide a foundation for interpreting parity-odd 4PCF.

Cosmology↗
DOE OSTI2026

Iterative HOMER with uncertainties

We present iHOMER, an iterative version of the HOMER method to extract Lund fragmentation functions from experimental data. Through iterations, we address the information gap between latent and observable phase spaces and systematically remove bias. To quantify uncertainties on the inferred weights, we use a combination of Bayesian neural networks and uncertainty-aware regression. We find that the combination of iterations and uncertainty quantification produces well-calibrated weights that accurately reproduce the data distribution. A parametric closure test shows that the iteratively learned fragmentation function is compatible with the true fragmentation function.

Butter, Anja [Heidelberg Univ. (Germany); Sorbonne↗
DOE OSTI2026

Neural posterior unfolding

Differential cross section measurements are the currency of scientific exchange in particle and nuclear physics. A key challenge for these analyses is the correction for detector distortions, known as deconvolution or unfolding. Binned unfolding of cross section measurements traditionally rely on the regularized inversion of the response matrix that represents the detector response, mapping pre-detector (`particle level') observables to post-detector (`detector level') observables. In this paper we introduce Neural Posterior Unfolding, a modern, Bayesian approach that leverages normalizing flows for unfolding. By using normalizing flows for neural posterior estimation, NPU offers several key advantages including implicit regularization through the neural network architecture, fast amortized inference that eliminates the need for repeated retraining, and direct access to the full uncertainty in the unfolded result. In addition to introducing NPU, we implement a classical Bayesian unfolding method called Fully Bayesian Unfolding (FBU) in modern Python so it can also be studied. These tools are validated on simple Gaussian examples and then tested on simulated jet substructure examples from the Large Hadron Collider (LHC). We find that the Bayesian methods are effective and worth additional development to be analysis ready for cross section measurements at the LHC and beyond.

Analysis and statistical methods↗
DOE OSTI2025

Inverse model based error detection in beamline optics

Optics tuning in transfer lines and LINACs can be challenging due to the fact that multiple combinations of machine settings can lead to the same diagnostic output. Moreover, the lack of a periodic solution can limit the ability to infer optics in the same way as rings from BPM signals. Model based approaches are often used to assist with the optics tuning in combination with optimization or parameter estimation. Here we have developed a novel approach using machine learning inverse models trained on a known configuration to detect variations in quadrupole settings without explicitly including them in the model. This paper shows a comparison of neural network models and linear models on both a simulation based study and experimental studies conducted at the AGS to RHIC transfer line at Brookhaven National Lab.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

On the impact of tidal generation and energy storage integration in PV-rich electric distribution systems

Deep decarbonization of power system operations requires the maximal utilization of available renewable resources. At distribution-level operations, however, grid operators can face numerous challenges in integrating renewables at scale owing to the inherent intermittence of renewable energy resources. These include phenomena such as voltage fluctuations, which are typically mitigated through control actuators such as on-load tap changers (OLTC) as well as energy storage devices, such as battery energy storage systems (BESS). On the one hand, high intermittence of the available renewable portfolio may require increasingly aggressive control of actuators, thereby accelerating the probability of equipment failure. On the other hand, integrating BESS operations and having a diverse renewable generation portfolio can typically help stagger power/energy flow to mitigate the aforementioned adverse impacts. In this paper, we employ a Bayesian framework for equipment lifetime estimation to understand the impact of including tidal energy resources and BESS in distribution system operations for feeders having substantial distribution photovoltaic generation. Our results indicate that while tidal energy alone may slightly decrease equipment reliability, the adverse impact on reliability is significantly magnified by a generation portfolio consisting of tidal generation and photovoltaic generation. Here, we also study the tidal and photovoltaic hosting capacity problem with and without energy storage systems using equipment reliability as an added constraint. We conclude that energy storage increases the reliability-constrained hosting capacity of the distribution system.

14 SOLAR ENERGY↗
DOE OSTI2025

Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors

Detectors in next-generation high-energy physics experiments face several daunting requirements, such as high data rates, damaging radiation exposure, and stringent constraints on power, space, and latency. To address these challenges, machine learning in readout electronics can be leveraged for smart detector designs, enabling intelligent inference and data reduction at-source. Variational autoencoders (VAEs) offer a variety of benefits for front-end readout; an on-sensor encoder can perform efficient lossy data compression while simultaneously providing a latent space representation that can be used for anomaly detection. Results are presented from low-latency and resource-efficient VAEs for front-end data processing in a futuristic silicon pixel detector. Encoder-based data compression is found to preserve good performance of off-detector analysis while significantly reducing the off-detector data rate as compared to a similarly sized data filtering approach. Furthermore, the latent space information is found to be a useful discriminator in the context of real-time sensor defect monitoring. Together, these results highlight the multifaceted utility of autoencoder-based front-end readout schemes and motivate their consideration in future detector designs.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Estimating ionization states and continuum lowering from ab initio path integral Monte Carlo simulations for warm dense hydrogen

Warm dense matter (WDM) is an active field of research, with applications ranging from astrophysics to inertial confinement fusion. Ionization degree and continuum lowering are important quantities to understand how materials behave under these conditions, but can be difficult to diagnose since experimental campaigns are limited and often require model-dependent analysis. This is especially true for hydrogen, which has a comparably low scattering cross section, making high-quality data particularly difficult to obtain. Consequently, building equation of state tables often relies on simulations in combination with untested approximations to extract properties from experiments. Here, we investigate an approach for extracting the ionization potential depression and ionization degree—quantities which are otherwise not directly accessible from the physical model—from first-principles path integral Monte Carlo (PIMC) simulations utilizing a chemical model. In contrast to experimental measurements, where noise and nonequilibrium effects add to the uncertainty of the inferred parameters, PIMC simulations provide a clean signal with well-defined thermodynamic conditions. Comparisons against commonly used models show a qualitative agreement, but we find deviations primarily for the high-density and high-temperature cases. We also demonstrate the decreasing sensitivity of the dynamic structure factor with respect to both ionization and continuum lowering for increasing scattering angles in x-ray Thomson scattering experiments. Our work has important implications for the design of future experiments, but also offers qualitative understanding of structure factors and the imaginary-time correlation function obtained from first-principles quantum Monte Carlo simulations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
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 OSTI2025

Paleoclimatic implications of glacial fluctuations in the Sierra Nevada del Cocuy, northern Andes, Colombia, during the Lateglacial and Holocene

The reconstruction of former mountain glaciers from geomorphic mapping and cosmogenic-nuclide surface-exposure dating provides a unique opportunity to infer patterns of past terrestrial climate variability. Tropical mountain glaciers are particularly valuable as there are comparatively few terrestrial climate proxies at equatorial latitudes relative to higher latitudes. As the single largest climate zone on Earth, the tropics play an outsized role in mediating global climate via the ocean-atmosphere transfer of latent heat and water vapor. Nonetheless, there remains a persistent gap in our understanding of how the tropics influenced – or were influenced by – the high-magnitude climate shifts of the Late Pleistocene, and whether this high-energy region simply responded to extratropical forcing or was itself a driver of global climatic change. To help address this knowledge gap, we analyzed geologic evidence for past glacial fluctuations in three adjacent valleys in the Sierra Nevada del Cocuy, the highest subrange of the Eastern Cordillera in the Colombian Andes, to provide a terrestrial record of atmospheric temperature during the latter part of Termination 1. Coupled with geomorphic mapping and paleo-snowline reconstructions, our beryllium-10 glacial chronology indicates that glaciers in the humid inner tropics underwent pronounced growth and gradual decay during the Antarctic Cold Reversal (14.5–12.8 ka) and Younger Dryas (12.8–11.7 ka) periods, respectively, following a trend that, according to directly dated moraine records from throughout both polar hemispheres, appears to have been global. While the specific mechanism(s) behind this large-scale behavior remains to be corroborated, we revisit the hypothesis that ocean-atmosphere heat transfer and water vapor flux are key drivers of abrupt Lateglacial temperature fluctuations. Subsequent to the Lateglacial, deglaciation of the Sierra Nevada del Cocuy accelerated during the Early Holocene, a pattern also observed in other tropical glacier records. More recently, the magnitude of snowline rise and glacier retreat over the last two centuries supports the view that modern tropospheric warming is anomalously strong at least relative to the last ∼16,000 years.

Andes↗
DOE OSTI2020

Comparison of Extended and Accelerated VOG Tests

The reprocessing of used nuclear fuel would release volatile radionuclides into the off-gas streams of a processing plant, including ¹²⁹I. The dissolver off-gas and more dilute vessel off-gas streams (VOG) must both be targeted when mitigating ¹²⁹I environmental release because each contains some amount of iodine. Iodine-129 will likely be found as elemental iodine (I₂) and methyl iodide (CH₃I) in these off-gas streams. Reduced silver-exchanged mordenite (AgZ or Ag⁰Z) has been investigated as a potential sorbent for iodine abatement and is studied here under prototypic VOG conditions. Because of the relatively low iodine abundance and high flow rates of the VOG, total iodine concentrations are expected to be in the parts per billion range. Thus, VOG experiments need to run for extended durations at low concentrations to reach sorbent saturation. Because the sorbent will be exposed to oxidizing gas streams for extended periods of time, aging effects and sorbent degradation need to be considered when designing a sorbent-based abatement system. Three sets of experiments were performed with the aim of determining how the adsorption of iodine by AgZ is affected by differing test durations and gas compositions. The first tested the capacity of sorbent aged for 8 months under a humid air stream. The second tested differences in sorbent capacity and mass transfer zone (MTZ) length during high-concentration (1200 ppb v ) CH₃I loading in a humid nitrogen gas (N₂) stream and a humid air gas stream over 28 days. The third tested sorbent capacity and MTZ length during low concentration (< 200 ppb v ) CH₃I and I₂ loading in a humid air stream over 9 months. The results of these tests suggest that AgZ capacity drops by ~50% after 1 month of aging, by ~60% after 2 months of aging, and then does not continue to decrease significantly at up to 8 months of aging. One-month aging tests that compared N₂ and air as the gas stream diluent resulted in similar maximum loading capacities and overall loading curves, indicating that the effects of aging cannot be mitigated by avoiding air as the balance gas. The 9-month extended VOG tests did not result in clear sorbent saturation at the inlet, but it can be inferred using an assumption of a maximum capacity of 45 mg I/g AgZ. Applying this assumption, then the MTZ of CH₃I is 12.8 cm, and the MTZ of I₂ is 12.4 cm. These results are similar to previous tests. Comparisons to tests completed at Idaho National Laboratory suggest that the CH₃I MTZ may be dependent on concentration. Future work should re-evaluate the design of the VOG iodine capture system with this updated data and should seek to understand fundamental characteristics of CH₃I adsorption by AgZ, such as the effect of concentration, gas velocity, and gas composition.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗