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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 307 records · Page 17

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Map-level baryonification: unified treatment of weak lensing two-point and higher-order statistics

Precision cosmology benefits from extracting maximal information from cosmic structures, motivating the use of higher-order statistics (HOS) at small spatial scales. However, predicting how baryonic processes modify matter statistics at these scales has been challenging. The baryonic correction model (BCM) addresses this by modifying dark-matter-only simulations to mimic baryonic effects, providing a flexible, simulation-based framework for predicting both two-point and HOS. We show that a 3-parameter version of the BCM can jointly fit weak lensing maps' two-point statistics, wavelet phase harmonics coefficients, scattering coefficients, and the third and fourth moments to within 2% accuracy across all scales ℓ < 2000 and tomographic bins for a DES-Y3-like redshift distribution ( z ≲ 2), using the FLAMINGO simulations. These results demonstrate the viability of BCM-assisted, simulation-based weak lensing inference of two-point and HOS, paving the way for robust cosmological constraints that fully exploit non-Gaussian information on small spatial scales.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2024

Ligand Many-Body Expansion as a General Approach for Accelerating Transition Metal Complex Discovery

Methods that accelerate the evaluation of molecular properties are essential for chemical discovery. While some degree of ligand additivity has been established for transition metal complexes, it is underutilized in asymmetric complexes, such as the square pyramidal coordination geometries highly relevant to catalysis. To develop predictive methods beyond simple additivity, we apply a many-body expansion to octahedral and square pyramidal complexes and introduce a correction based on adjacent ligands (i.e., the cis interaction model). We first test the cis interaction model on adiabatic spin-splitting energies of octahedral Fe(II) complexes, predicting DFT-calculated values of unseen binary complexes to within an average of 1.4 kcal/mol. Uncertainty analysis reveals the optimal basis, comprising the homoleptic and mer symmetric complexes. We next show that the cis model (i.e., the cis interaction model solved for the optimal basis) infers both DFT- and CCSD(T)-calculated model catalytic reaction energies to within 1 kcal/mol on average. The cis model predicts low-symmetry complexes with reaction energies outside the range of binary complex reaction energies. We observe that trans interactions are unnecessary for most monodentate systems but can be important for some combinations of ligands, such as complexes containing a mixture of bidentate and monodentate ligands. Lastly, we demonstrate that the cis model may be combined with Δ-learning to predict CCSD(T) reaction energies from exhaustively calculated DFT reaction energies and the same fraction of CCSD(T) reaction energies needed for the cis model, achieving around 30% of the error from using the CCSD(T) reaction energies in the cis model alone.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2026

Generative modeling enables molecular structure retrieval from Coulomb explosion imaging

Capturing the structural changes that molecules undergo during chemical reactions in real space and time is a long-standing dream and an essential prerequisite for understanding and ultimately controlling femtochemistry. A key approach to tackle this challenging task is Coulomb explosion imaging, which benefited decisively from recently emerging high-repetition-rate X-ray free-electron laser sources. With this technique, information on the molecular structure is inferred from the momentum distributions of the ions produced by the rapid Coulomb explosion of molecules. Retrieving molecular structures from these distributions poses a highly non-linear inverse problem that remains unsolved for molecules consisting of more than a few atoms. Here, we address this challenge using a diffusion-based Transformer neural network. We show that the network reconstructs unknown molecular geometries from ion-momentum distributions with a mean absolute error below one Bohr radius, which is half the length of a typical chemical bond.

Artificial Intelligence (cs.AI)↗

DOE OSTI2024

Toward Accelerated Nuclear-physics Parameter Estimation from Binary Neutron Star Mergers: Emulators for the Tolman–Oppenheimer–Volkoff Equations

Abstract Gravitational-wave observations of binary neutron-star (BNS) mergers have the potential to revolutionize our understanding of the nuclear equation of state (EOS) and the fundamental interactions that determine its properties. However, Bayesian parameter estimation frameworks do not typically sample over microscopic nuclear-physics parameters that determine the EOS. One of the major hurdles in doing so is the computational cost involved in solving the neutron-star structure equations, known as the Tolman–Oppenheimer–Volkoff (TOV) equations. In this paper, we explore approaches to emulating solutions for the TOV equations: multilayer perceptrons (MLPs), Gaussian processes, and a data-driven variant of the reduced basis method (RBM). We implement these emulators for three different parameterizations of the nuclear EOS, each with a different degree of complexity represented by the number of model parameters. We find that our MLP-based emulators are generally more accurate than the other two algorithms, whereas the RBM results in the largest speedup with respect to the full high-fidelity TOV solver. We employ these emulators for a simple parameter inference using a potentially loud BNS observation and show that the posteriors predicted by our emulators are in excellent agreement with those obtained from the full TOV solver.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2026

A Sample of Nearby Isolated Dwarf Galaxies: A First Look at the Mass Function of Field Dwarfs

We present the results of the Exploration of Local VolumE Survey-Field, a survey of the dwarf galaxies in the Local Volume (LV; D < 10 Mpc) over roughly 3000 deg 2 , focusing on the field dwarf population. Candidates are detected using a semiautomated algorithm tailored for low-surface brightness dwarfs. Using tests with injected galaxies, we show the detection is 50% complete to m g ∼ 20 mag and M ⋆ ∼ 10 6 M ⊙ . Candidates are confirmed to be true nearby dwarfs through distance measurements including redshift, tip of the red giant branch, and surface brightness fluctuations. We identify isolated field dwarfs using various environmental criteria. Over the survey footprint, we detect and confirm 95 LV dwarfs, 44 of which we consider isolated. Using this sample, we infer the field dwarf mass function and find good agreement at the high-mass end with previous redshift surveys and with the predictions of the Illustris-TNG simulation. This sample of isolated field dwarfs represents a powerful dataset to investigate aspects of small-scale structure and the effect of environment on dwarf galaxy evolution.

Carlsten, Scott G. [Princeton Univ., NJ (United St↗

DOE OSTI2025

Testing T2K’s Bayesian constraints with priors in alternate parameterisations

Bayesian analysis results require a choice of prior distribution. In long-baseline neutrino oscillation physics, the usual parameterisation of the mixing matrix induces a prior that privileges certain neutrino mass and flavour state symmetries. Here we study the effect of privileging alternate symmetries on the results of the T2K experiment. We find that constraints on the level of CP violation (as given by the Jarlskog invariant) are robust under the choices of prior considered in the analysis. On the other hand, the degree of octant preference for the atmospheric angle depends on which symmetry has been privileged.

Bayesian Inference↗

DOE OSTI2026

Longitudinal Phase Space Tomography for the Booster Synchrotron

Efforts in the study of the longitudinal behavior of charged particles in the Fermilab Booster can be catalyzed with an image of the two-dimensional phase space distribution. In the past, tomography has been extensively employed in the reconstruction of the phase space in accelerators such as the Recycler at Fermilab and the Proton Synchrotron Booster at CERN. However, such a capability had yet to realize for the Fermilab Booster synchrotron. In this work, the first successful tomographic phase space reconstruction of a low-energy Booster bunch is presented along with validation metrics. A numerical turn-by-turn model of the longitudinal particle dynamics in the Booster has been implemented, which utilizes a fast, map-based particle transport algorithm. Using a sinogram generated from the Wall Current Monitor, the iterative reconstruction algorithm recovers a discretized image of the original phase space distribution at variable resolution. The reconstruction result shows low root-mean-square error and a rapid convergence toward the solution, providing strong evidence of accuracy. Future and ongoing work includes modeling high-energy bunches above transition and using tomography to infer certain machine parameters such as synchronous phase, peak gap voltage, and synchronous energy in addition to the phase space distribution.

Ebeid, Safi [Unlisted]↗

DOE OSTI2026

Longitudinal Phase Space Tomography for the Booster Synchrotron (Abstract Only)

Ebeid, Safi [Unlisted, US]↗

DOE OSTI2026

Artificial Intelligence for Event Reconstruction and Higgs Physics at CMS and Future Colliders

This dissertation charts a trajectory in which advances in artificial intelligence (AI) play a central role in pushing the high-energy physics frontier, complementing progress driven by higher collision energies and larger colliders. The discovery potential of the LHC and future colliders relies on accurate reconstruction of increasingly complex particle collision events. In the CMS experiment, this task is performed by the particle-flow (PF) algorithm. This dissertation presents the first implementation of a machine-learning-based particle-flow (MLPF) reconstruction in the CMS detector based on transformer architectures. In simulated top quark--antiquark pair (ttbar) events under LHC Run~3 (2023--2024) conditions, MLPF improves jet energy resolution by 10--20\% compared to standard PF for jets with transverse momentum between 30--100\GeV. Runtime performance is evaluated using simulated multijet events, with a median inference time of 20\unit{ms} per event on an NVIDIA L4 GPU, compa red to approximately 110\unit{ms} for standard PF. The MLPF algorithm is also validated on Run~3 collision data, representing the first data-validated ML-based reconstruction pipeline at any LHC experiment. We then extend MLPF toward future electron--positron colliders and introduce the first full-simulation cross-detector transfer learning workflow for PF reconstruction. The model is pre-trained on simulated events from the Compact Linear Collider detector (CLICdet) and fine-tuned on the CLIC-like detector (CLD) proposed for the Future Circular Collider (FCC). This approach achieves up to a 40\% improvement in jet energy resolution over rule-based reconstruction while reducing the required training dataset size by an order of magnitude, demonstrating the potential of AI to accelerate detector development and optimization. This dissertation also demonstrates how modern AI techniques enhance the sensitivity of LHC physics analyses. A CMS search for highly Lorentz-boosted Higgs bosons decaying to \textrm{W} boson pairs is presented, focusing on the single-lepton final state. A dedicated fine-tuning strategy for \ParT yields an approximately 70\% increase in expected sensitivity relative to the baseline model. The analysis uses proton--proton collision data at a center-of-mass energy of \ensuremath{\sqrt{s}=13\TeV} collected by CMS between 2016 and 2018, corresponding to an integrated luminosity of 138\ensuremath{\ \mathrm{fb}^{-1}}. The expected significance of the search is $1.86\sigma$, with an observed signal strength of $-0.19^{+0.48}_{-0.46}$. Finally, explainable AI techniques are applied to the MLPF and \ParticleNet algorithms using layerwise relevance propagation, showing that both models base their predictions on physically meaningful features consistent with our physics intuition. Together, these results demonstrate how advanced AI methods can enhance reconstruction, analysis sensitivity, and interpretability, shaping the next era of experimental parti cle physics.

Mokhtar, Farouk [UC, San Diego]↗

DOE OSTI2026

Longitudinal Phase Space Tomography for the Booster Synchrotron

Efforts in the study of the longitudinal behavior of charged particles in the Fermilab Booster can be catalyzed with an image of the two-dimensional phase space distribution. In the past, tomography had been employed in the reconstruction of the phase space in accelerators such as the Recycler at Fermilab and the Proton Synchrotron Booster at CERN. However, such a capability had yet to realize for the Fermilab Booster synchrotron. In this work, the first successful tomographic phase space reconstruction of a low-energy Booster bunch is presented along with validation metrics. A numerical turn-by-turn model of the longitudinal particle dynamics in the Booster has been implemented, which utilizes a fast, map-based particle transport algorithm. Using a sinogram generated from the Wall Current Monitor signal, the iterative reconstruction algorithm recovers a discretized image of the original phase space distribution at variable resolution. The reconstruction result shows low root-mean-square error and a rapid convergence toward the solution, providing strong evidence of accuracy. Future and ongoing work includes modeling high-energy bunches above transition and using tomography to infer certain machine parameters such as synchronous phase, peak gap voltage, and synchronous energy in addition to the phase space distribution.

Ebeid, Safi [Unlisted; Fermilab]↗

DOE OSTI2025

Two-Step Procedure to Detect Cosmological Gravitational Wave Backgrounds with Next-Generation Terrestrial Gravitational-Wave Detectors

Cosmological gravitational-wave backgrounds are an exciting science target for next-generation ground-based detectors, as they encode invaluable information about the primordial Universe. However, any such background is expected to be obscured by the astrophysical foreground from compact-binary coalescences. We propose a novel framework to detect a cosmological gravitational-wave background in the presence of binary black holes and binary neutron star signals with next-generation ground-based detectors, including Cosmic Explorer and the Einstein Telescope. Our procedure involves first removing all the individually resolved binary black hole signals by notching them out in the time-frequency domain. Then, we perform joint Bayesian inference on the individually resolved binary neutron star signals, the unresolved binary neutron star foreground, and the cosmological background. For a flat cosmological background, we find that we can claim detection at 5⁢𝜎 level when Ω ref ≥ 2.7 × 10 −12 /$\sqrt{𝑇_{obs}/yr}$, where 𝑇 obs is the observation time (in years), which is within a factor of ≲ 2 from the sensitivity reached in the absence of these astrophysical foregrounds.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2024

Impact of Varying Dark Energy on Future Large-Scale Structure Studies

The cosmos withholds multiple mysteries such as dark forms of matter and energy that are yet beyond human comprehension. In the standard cosmological model, $\Lambda$CDM, the cosmological constant $\Lambda$ is thought to be responsible for the late accelerated expansion of the Universe. However, recent results from the Dark Energy Spectroscopic Instrument (DESI) suggest the possibility of evolving dark energy, which warrants further exploration. Future surveys such as the Rubin Observatory Legacy Survey of Space and Time (LSST) will map the large-scale structure (LSS) with unprecedented precision, giving us valuable statistical information about the cosmos. The goal of this research is to investigate the potential impact of an evolving dark energy scenario on cosmological parameters constrained by LSS probes, as will be mapped by the LSST. For this, we examine the power spectra of lens galaxies, source galaxies, and cross-power spectra between lens galaxies and source galaxies. The combination of these statistics is commonly referred to as '3 $\times$ 2 points'. For this investigation, we created a set of simulations resembling LSST data and used them to perform cosmological parameter inference in two scenarios: one in which the simulated data is based on the fiducial model and another on evolving dark energy. We then examined the degeneracy between the cosmological parameters and checked for potential shifts in the parametric space when the data contains dynamical dark energy but the modeling assumes $\Lambda$CDM. Our findings indicate that mismodeling the dark energy equation of state can significantly impact parameter inference, particularly affecting the total matter density, $\Omega_m$, and the growth of structures, as represented by the $S_8$ parameter. These results highlight the importance of further exploring extensions of the $\Lambda$CDM model in future LSS studies.

Yaman Acharya, A.↗

DOE OSTI2020

Detection of very-high-energy gamma-ray transients with monitoring facilities

ABSTRACT The observation of very-high-energy γ-rays (VHE γ-rays, $E \gt 100\,$ GeV) has ledto the identification of extremely energetic processes and particle-acceleration sites both within our Galaxy and beyond. It is expected that VHE facilities, such as the Cherenkov Telescope Array, will explore these sources with an unprecedented level of detail. However, the transient and unpredictable nature of many important processes means that their observation requires the development of proper monitoring strategies. In this study, we estimate the properties of VHE transients that can be effectively detected by monitoring facilities. We use data collected by the Fermi-LAT instrument during its monitoring campaign to select events that are probably associated with VHE emission. We use this sample to estimate the frequency, the luminosity and the time-scales of various transients, focusing on blazar flares and gamma-ray bursts. We discuss how the balance between the field of view, sensitivity and duty cycle of an observatory affects the likelihood of detecting transients that occur at the inferred rates, and we conclude by describing the contribution that current and near-future monitoring facilities can make to the identification and study of VHE transient emission.

La Mura, G.↗

DOE OSTI2021

A Reconfigurable Neural Network ASIC for Detector Front-End Data Compression at the HL-LHC

Despite advances in the programmable logic capabilities of modern trigger systems, a significant bottleneck remains in the amount of data to be transported from the detector to off-detector logic where trigger decisions are made. We demonstrate that a neural network (NN) autoencoder model can be implemented in a radiation-tolerant application-specific integrated circuit (ASIC) to perform lossy data compression alleviating the data transmission problem while preserving critical information of the detector energy profile. For our application, we consider the high-granularity calorimeter from the Compact Muon Solenoid (CMS) experiment at the CERN Large Hadron Collider. The advantage of the machine learning approach is in the flexibility and configurability of the algorithm. By changing the NN weights, a unique data compression algorithm can be deployed for each sensor in different detector regions and changing detector or collider conditions. To meet area, performance, and power constraints, we perform quantization-aware training to create an optimized NN hardware implementation. The design is achieved through the use of high-level synthesis tools and the hls4ml framework and was processed through synthesis and physical layout flows based on a low-power (LP)-CMOS 65-nm technology node. The flow anticipates 200 Mrad of ionizing radiation to select gates and reports a total area of 3.6 mm 2 and consumes 95 mW of power. The simulated energy consumption per inference is 2.4 nJ. Furthermore, this is the first radiation-tolerant on-detector ASIC implementation of an NN that has been designed for particle physics applications.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

Announcing the Biomedical Data Translator: Initial Public Release

ABSTRACT The growing availability of biomedical data offers vast potential to improve human health, but the complexity and lack of integration of these datasets often limit their utility. To address this, the Biomedical Data Translator Consortium has developed an open‐source knowledge graph–based system—Translator—designed to integrate, harmonize, and make inferences over diverse biomedical data sources. We announce here Translator's initial public release and provide an overview of its architecture, standards, user interface, and core features. Translator employs a scalable, federated, knowledge graph framework for the integration of clinical, genomic, pharmacological, and other biomedical knowledge sources, enabling query retrieval, inference, and hypothesis generation. Translator's user interface is designed to support the exploration of knowledge relationships and the generation of insights, without requiring deep technical expertise and gradually revealing more detailed evidence, provenance, and confidence information, as needed by a given user. To demonstrate Translator's application and impact, we highlight features of the user interface in the context of three real‐world use cases: suggesting potential therapeutics for patients with rare disease; explaining the mechanism of action of a pipeline drug; and screening and validating drug candidates in a model organism. We discuss strengths and limitations of reasoning within a largely federated system and the need for rich concept modeling and deep provenance tracking. Finally, we outline future directions for enhancing Translator's functionality and expanding its data sources. Translator represents a significant step forward in making complex biomedical knowledge more accessible and actionable, aiming to accelerate translational research and improve patient care.

Research & Experimental Medicine↗

DOE OSTI2021

Surveying Grassland Islands: the genetics and performance of Appalachian switchgrass (Panicum virgatum L.) collections

The interior Southeastern United States could contain novel germplasm for the bioenergy crop switchgrass due to its diverse habitats and geographic location between genetic subpopulations (Atlantic, Midwest, and Gulf). Collections from this region could accelerate breeding progress, contribute to conservation efforts, and improve understanding of isolated grasslands in the region. This study located 22 sites in the Midsouth region and obtained 1,521,210 single nucleotide polymorphism markers of 202 individuals through genotype-by-sequencing. Here, individuals were evaluated for flowering time, winter survival and tiller number. Comparison to a national diversity panel revealed that branches of two major subpopulations occur in the region with two levels of polyploidy: Atlantic tetraploids and Midwest octoploids. Two locations contained admixed octoploid individuals with Midwest and Gulf genetics. Field performance of the Midwest octoploids conformed with prior reported performance of the Midwest subpopulation, although three sites contained promising late flowering traits. The Atlantic tetraploids had moderate winter survival, short stature, and anomalously early flowering. Atlantic populations mostly occurred in marginal sites and their morphological and flowering time adaptations may be a resource conservation strategy. Demographic inference of historical effective population size variation in a subset of tetraploid locations indicated a widespread recent decline in effective population size. This pattern is consistent with isolation of these switchgrass communities from larger populations and is further supported by evidence of inbreeding within the populations (F I = 0.18). The populations documented in this study contain novel genetic diversity and adaptations to a range of marginal habitats. Therefore, this study provides a new source of germplasm for future breeding and conservation programs.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

Agentic framework for programmatic crystal structure generation using a fine-tuned worker–supervisor large language model

Platinum group metals (PGMs) underpin many catalytic technologies but face severe supply constraints, motivating the search for alternative materials and computational methods to accelerate discovery. While atomistic simulation tools such as Pymatgen and ASE have streamlined structure manipulation, they require detailed inputs, limiting accessibility for experimentalists and slowing early-stage exploration. Here, in this study, we present an AI-driven agentic framework that orchestrates worker–supervisor large language models (LLMs). The worker translates natural-language prompts of varying abstraction into valid crystallographic structures using a compact LLM fine-tuned with low-rank adaptation on a curated text–code–CIF dataset, emphasizing energy-efficient training. Benchmarking against the baseline CodeGen-350M-mono model shows that fine-tuning reduces hallucination rates from 100% to as low as 5% and improves structural match accuracy to up to 82% for fully specified inputs. Accuracy declines with decreasing prompt detail but remains nontrivial even when only stoichiometry and space group are provided, underscoring the LLM’s capacity for crystallographic inference. The supervisor Claude LLM evaluates the outputs and triggers iterative refinement through the worker’s built-in structure manipulation capabilities (e.g., supercell scaling, strain, vacancy, and substitution operations). We further demonstrate use cases for technologically relevant catalysts, including IrO 2 , pyrochlore Pb 2 Ir 2 O 7 , Ni 2 FeO 4 , and Ni 3 Mo, where the framework generates physically consistent structures that can be refined via geometry optimization. This work introduces a low-energy, language-driven pathway for integrating human and machine intelligence in materials design, paving the way for AI-assisted synthesis planning and high-throughput screening of complex oxides.

AI agent↗

DOE OSTI2025

A physics-constrained neural ordinary differential equations approach for robust learning of stiff chemical kinetics

The high computational cost associated with solving for detailed chemistry poses a significant challenge for predictive computational fluid dynamics (CFD) simulations of turbulent reacting flows. While deep learning techniques have been explored to develop faster surrogate models, they often fail to integrate reliably with CFD solvers. This instability arises because traditional deep learning approaches optimize for training error without ensuring compatibility with ordinary differential equation (ODE) solvers, resulting in accumulation of errors over time. Recently, neuralODE (NODE) based approaches have been shown to be a promising technique to emulate and accelerate detailed chemistry computations. Here, in the present work, we extend this NODE framework for stiff chemical kinetics by incorporating mass conservation constraints directly into the loss function during training. This ensures that the total mass as well as the individual elemental species masses are conserved in an a-posteriori manner. Proof-of-concept studies are performed with the novel physics-constrained NODE (PC-NODE) approach for homogeneous autoignition of hydrogen-air mixture over a range of composition and thermodynamic conditions. It is demonstrated that the PC-NODE framework not only improves the physical consistency of the resulting data-driven model with respect to mass conservation criteria, but also improves training efficiency. PC-NODE is shown to achieve 2–100× speedup relative to the hydrogen-air detailed chemical mechanism depending on the type of the ODE solver (implicit or explicit) used during autoregressive inference tests. Lastly, a-posteriori studies are performed wherein the trained PC-NODE model is coupled with a CFD solver. It is shown that higher accuracy is achieved with PC-NODE relative to the purely data-driven NODE approach. Moreover, PC-NODE also exhibits robustness and generalizability to unseen initial conditions from within (interpolative capability) as well as outside (extrapolative capability) the training regime.

computational combustion↗