Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “network acceleration”

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 1,243 records · Page 69
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Machine learning enhanced predictions of ICRF heating: Overcoming numerical limitations via data curation

In this work, we present the development of robust surrogate models for Ion Cyclotron Range of Frequencies (ICRF) and High-Harmonic Fast Wave (HHFW) heating predictions in fusion plasmas. Building upon our previous efforts to achieve real-time capable models, we identify the cause of the outliers found using TORIC in certain HHFW heating scenarios. The outliers are observed to be spurious ion Bernstein wave (IBW)-like modes caused by a wavelength control algorithm designed to address challenging scenarios with high perpendicular wavenumbers. The effect arises from the modulation in the perpendicular susceptibility, which can induce sign reversal and IBW-like propagation for scenarios featuring normalized ion Larmor radius λ i ≫ 1. We use TORIC with this algorithm disabled to generate a novel HHFW-NSTX database that is free of outliers. Surrogate models trained on this database, including Random Forest Regressor (RFR), Multi-Layer Perceptrons, and Gaussian Process Regressors (GPR), demonstrate the ability to accurately predict HHFW heating profiles, with regression scores of R 2 ∈[0.93−0.99]. Additionally we demonstrate that it is possible to generalize predictions beyond training data by the use of both RFR and GPR models, enabling the prediction of scenarios previously limited to the original model. GPR models also provide uncertainty quantification, offering insights into model confidence. This work introduces a comprehensive Verification, Validation, and Uncertainty Quantification methodology for surrogate modeling, applicable not only to ICRF heating but also to other RF heating challenges and fusion physics problems. Beyond accelerated inference, these models show effective extrapolation capabilities, providing an alternative for addressing numerical challenges.

Artificial neural networks↗
DOE OSTI2025

Temperature-dependent solid electrolyte interphase reactions drive performance in lithium-mediated nitrogen reduction to ammonia

The solid electrolyte interphase (SEI) is a vital component to control mass transport and selectivity in the lithium-mediated reduction of N 2 to NH 3 (Li-N 2 R). Finding strategies that generate the optimal SEI, a complex network of organic and inorganic species, can potentially improve Li-N 2 R performance. Here, we unravel structure-property relationships of the SEI by correlating its composition with the NH 3 faradaic efficiency (FENH3). By modifying the reaction temperature, we alter electrolyte decomposition reactions and observe changes in the SEI that explain FE NH3 trends between electrolyte solvents. We quantify a complex reaction environment at elevated temperatures where SEI formation is counteracted by etching reactions. This tradeoff leads to temporal fluctuations of FE NH3 , but the maximal FE NH3 can reach up to 40%, the highest value reported for batch cells at ambient pressure, thus far. In conclusion, our work underscores the potential of novel electrolytes that steer SEI selectivity and, ultimately, improve Li-N 2 R performance.

electrocatalytic nitrogen reduction↗
DOE OSTI2025

Monitoring the Structural Health of the Stage-Four Gibbs Resistor In Order to Maintain a Functioning Pulse-Forming Network

The Dual-Axis Radiographic Hydrodynamic Test (DARHT) facility at Los Alamos National Laboratory (LANL) is a critical facility used for nuclear weapons research and development (Los Alamos National Laboratory). Its primary function is to provide high-resolution, real-time images of the behavior of materials under extreme conditions, specifically during the hydrodynamic testing of nuclear weapons surrogates. The facility uses advanced radiographic techniques, such as dual-axis X-ray imaging, to capture detailed snapshots of these materials as they react to high-pressure environments. DARHT plays a key role in maintaining the safety, security, and reliability of the U.S. nuclear arsenal, supporting the Stockpile Stewardship Program. The facility helps ensure that nuclear weapons perform as designed without the need for nuclear tests. Its dual-axis radiography provides more precise data than traditional single-axis imaging, offering a comprehensive view of the internal dynamics of a weapon's primary stage.

42 ENGINEERING↗
DOE OSTI2023

Data-driven predictions of the time remaining until critical global warming thresholds are reached

Leveraging artificial neural networks (ANNs) trained on climate model output, we use the spatial pattern of historical temperature observations to predict the time until critical global warming thresholds are reached. Although no observations are used during the training, validation, or testing, the ANNs accurately predict the timing of historical global warming from maps of historical annual temperature. The central estimate for the 1.5 °C global warming threshold is between 2033 and 2035, including a ±1σ range of 2028 to 2039 in the Intermediate (SSP2-4.5) climate forcing scenario, consistent with previous assessments. However, our data-driven approach also suggests a substantial probability of exceeding the 2 °C threshold even in the Low (SSP1-2.6) climate forcing scenario. While there are limitations to our approach, our results suggest a higher likelihood of reaching 2 °C in the Low scenario than indicated in some previous assessments—though the possibility that 2 °C could be avoided is not ruled out. Explainable AI methods reveal that the ANNs focus on particular geographic regions to predict the time until the global threshold is reached. Our framework provides a unique, data-driven approach for quantifying the signal of climate change in historical observations and for constraining the uncertainty in climate model projections. Given the substantial existing evidence of accelerating risks to natural and human systems at 1.5 °C and 2 °C, our results provide further evidence for high-impact climate change over the next three decades.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Accelerating accurate simulations of calorimeter showers with normalizing flows and probability density distillation

Recently, we introduced caloflow, a high-fidelity generative model for GEANT4 calorimeter shower emulation based on normalizing flows. Here, we present caloflow v2, an improvement on our original framework that speeds up shower generation by a further factor of 500 relative to the original. The improvement is based on a technique called probability density distillation, originally developed for speech synthesis in the machine learning literature, and which we develop further by introducing a set of powerful new loss terms. We demonstrate that caloflow v2 preserves the same high fidelity as the original using qualitative (average images, histograms of high-level features) and quantitative (classifier metric between geant4 and generated samples) measures. The result is a generative model for calorimeter showers that matches the state of the art in speed (a factor of 10 4 faster than GEANT4) and greatly surpasses the previous state of the art in fidelity.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Semi-supervised graph neural networks for pileup noise removal

Abstract The high instantaneous luminosity of the CERN Large Hadron Collider leads to multiple proton–proton interactions in the same or nearby bunch crossings (pileup). Advanced pileup mitigation algorithms are designed to remove this noise from pileup particles and improve the performance of crucial physics observables. This study implements a semi-supervised graph neural network for particle-level pileup noise removal, by identifying individual particles produced from pileup. The graph neural network is firstly trained on charged particles with known labels, which can be obtained from detector measurements on data or simulation, and then inferred on neutral particles for which such labels are missing. This semi-supervised approach does not depend on the neutral particle pileup label information from simulation, and thus allows us to perform training directly on experimental data. The performance of this approach is found to be consistently better than widely-used domain algorithms and comparable to the fully-supervised training using simulation truth information. The study serves as the first attempt at applying semi-supervised learning techniques to pileup mitigation, and opens up a new direction of fully data-driven machine learning pileup mitigation studies.

43 PARTICLE ACCELERATORS↗
NASA NTRS2009

Land-Use and Land-Cover Change around Mobile Bay, Alabama from 1974-2008

This document summarizes the major findings of a Gulf of Mexico Application Pilot project led by NASA Stennis Space Center (SSC) in conjunction with a regional collaboration network of the Gulf of Mexico Alliance (GOMA). NASA researchers processed and analyzed multi-temporal Landsat data to assess land-use and land-cover (LULC) changes in the coastal counties of Mobile and Baldwin, AL between 1974 and 2008. Our goal was to create satellite-based LULC data products using methods that could be transferable to other coastal areas of concern within the Gulf of Mexico. The Mobile Bay National Estuary Program (MBNEP) is the primary end-user, however, several other state and local groups may benefit from the project s data products that will be available through NOAA-NCDDC s Regional Ecosystem Data Management program. Mobile Bay is a critical ecologic and economic region in the Gulf of Mexico and to the entire country. Mobile Bay was designated as an estuary of national significance in 1996. This estuary receives the fourth largest freshwater inflow in the United States. It provides vital nursery habitat for commercially and recreationally important fish species. It has exceptional aquatic and terrestrial bio-diversity, however, its estuary health is influenced by changing LULC patterns, such as urbanization. Mobile and Baldwin counties have experienced a population growth of 1.1% and 20.5% from 2000-2006. Urban expansion and population growth are likely to accelerate with the construction and operation of the ThyssenKrupp steel mill in the northeast portion of Mobile County. Land-use and land-cover change can negatively impact Gulf coast water quality and ecological resources. The conversion of forest to urban cover types impacts the carbon cycle and increases the freshwater and sediment in coastal waters. Increased freshwater runoff decreases salinity and increases the turbidity of coastal waters, thus impacting the growth potential of submerged aquatic vegetation (SAV), which is critical nursing ground for many Gulf fish species. A survey of Mobile Bay SAV showed widespread decreases since the 1940s. Prior to our project, coastal environmental managers in Baldwin and Mobile counties needed more understanding of the historical LULC for properly assessing the impacts of urbanization. In particular, more information on the location and extent of changing urbanization LULC patterns was needed to aid LULC planning and to assess predictions of future LULC patterns. Our products will assist the coastal environmental managers and land-use planners in making better community growth planning decisions. Our project also will help to establish a historical baseline of LULC distributions, which is a fundamental need in any stewardship plan. The primary research objective of our project was to produce historic and current geospatial LULC change products across a 34-year time frame. A multi-decadal coastal LULC change product was the major project deliverable. The geographic extent and nature of change was quantified and assessed for the upland herbaceous, barren, open water, urban, upland forest, woody wetland, and non-woody wetlanddominated land cover types. We focused on regional analyses of decadal-scale urban expansion and watershed-scaled analyses of LULC change for multiple areas of concern to the Mobile Bay NEP (Figure A). We used the following dates to derive LULC classification products from Landsat data: 1974, 1979, 1984, 1988, 1991, 1996, 2001, 2005, and 2008. We assessed the accuracy of our products using randomly sampled locations and digital geospatial reference data including field survey data, high resolution orthorectified aerial photography, high resolution multispectral and panchromatic satellite data displays (from QuickBird and Corona sensors), digital elevation model data, and National Wetlands Inventory wetland cover type data. NOAA s Coastal Change Assessment Program s (C-CAP) and National Land Cover Database (NLCD) procts were used for qualitative comparison in assessing map accuracy. We calculated an average overall classification accuracy of 87% with similar overall accuracies for the older (MSS) and newer (TM and ETM) Landsat LULC products.

Ellis, Jean↗
NASA NTRS2014

Using SpF to Achieve Petascale for Legacy Pseudospectral Applications

Pseudospectral (PS) methods possess a number of characteristics (e.g., efficiency, accuracy, natural boundary conditions) that are extremely desirable for dynamo models. Unfortunately, dynamo models based upon PS methods face a number of daunting challenges, which include exposing additional parallelism, leveraging hardware accelerators, exploiting hybrid parallelism, and improving the scalability of global memory transposes. Although these issues are a concern for most models, solutions for PS methods tend to require far more pervasive changes to underlying data and control structures. Further, improvements in performance in one model are difficult to transfer to other models, resulting in significant duplication of effort across the research community. We have developed an extensible software framework for pseudospectral methods called SpF that is intended to enable extreme scalability and optimal performance. Highlevel abstractions provided by SpF unburden applications of the responsibility of managing domain decomposition and load balance while reducing the changes in code required to adapt to new computing architectures. The key design concept in SpF is that each phase of the numerical calculation is partitioned into disjoint numerical kernels that can be performed entirely inprocessor. The granularity of domain decomposition provided by SpF is only constrained by the datalocality requirements of these kernels. SpF builds on top of optimized vendor libraries for common numerical operations such as transforms, matrix solvers, etc., but can also be configured to use open source alternatives for portability. SpF includes several alternative schemes for global data redistribution and is expected to serve as an ideal testbed for further research into optimal approaches for different network architectures. In this presentation, we will describe our experience in porting legacy pseudospectral models, MoSST and DYNAMO, to use SpF as well as present preliminary performance results provided by the improved scalability.

DYNAMO↗
NASA NTRS2022

The Sun Radio Interferometer Space Experiment (SunRISE) Mission

The Sun Radio Interferometer Space Experiment (SunRISE) will provide an entirely new view on particle acceleration and transport in the inner heliosphere by creating the first low radio frequency interferometer in space to localize heliospheric radio emissions. By imaging and determining the location of decametric-hectometric (DH, 0.1 MHz–23 MHz) solar radio bursts, SunRISE will provide key information on particle acceleration mechanisms associated with coronal mass ejections (CMEs) and the magnetic field topology from active regions into interplanetary space. The SunRISE Observatory will consist of six space vehicles in a passive formation, in orbits designed to keep them within approximately 10 km of each other, and flying in a supersynchronous geosynchronous Earth orbit (GEO). Each space vehicle consists of a Solar DH-GNSS payload and a 6U form factor spacecraft. The SunRISE Observatory together with significant ground-based processing, will enable imaging of the Sun in a portion of the spectrum that is blocked by the ionosphere and cannot be observed from Earth. Key aspects that enable this mission are that only position knowledge of the space vehicles is required, not active control, and that the architecture involves a modest amount of on-board processing coupled with significant ground-based processing for navigation, position determination, and science operations. Mission-enabling advances in software-defined radios, GPS navigation and timing, and small spacecraft technologies, developed and flown on the DARPA High Frequency Research (DHFR) and the Community Initiative for Continuing Earth Radio Occultation (CICERO) have made this mission affordable and low-risk. The SunRISE mission will exploit the multiple spacecraft per aperture (MSPA) capability of NASA’s Deep Space Network (DSN), for more efficient data transfers of larger data volumes, and utilize commercial access to space, in which the SunRISE space vehicles will be carried to their target orbit as secondary payloads in conjunction with a larger host spacecraft intended for GEO.

Neilson, Tim↗
NASA NTRS2014

Public-Private Collaborations with Earth-Space Benefits

The NASA Human Health and Performance Center (NHHPC) was established in October 2010 to promote collaborative problem solving and project development to advance human health and performance innovations benefiting life in space and on Earth. The NHHPC, which now boasts over 135 corporate, government, academic and non-profit members, has convened four successful workshops and engaged in multiple collaborative projects. The center is currently developing a streamlined partner engagement process to capture technical needs and opportunities of NHHPC members, facilitate partnership development, and establish and manage collaborative projects for NASA. The virtual center facilitates member engagement through a variety of vehicles, including annual inperson workshops, webcasts, quarterly electronic newsletters, web postings, and the new system for partner engagement. The most recent NHHPC workshop was conducted in November 2013 on the topic of "Accelerating Innovation: New Organizational Business Models," and focused on various collaborative approaches successfully used by organizations to achieve their goals. The powerful notion of collaboration across sectors to solve intractable problems was recently highlighted in Williams Eggers' book "The Solution Revolution,"i which provides numerous examples of how business, government and social enterprises partner to solve tough problems. Mr. Eggers was a keynote speaker at the workshop, along with Harvard Business School, Jump Associates, and the Conrad Foundation. The robust program also included an expert panel addressing collaboration across sectors, four interactive breakout sessions, and a concluding keynote on innovative ways to increase science, technology, engineering, and math (STEM) education by NASA Associate Administrator for Education, Leland Melvin. The NHHPC forum also provides a platform for international partners to interact on many topics. Members from around the world include ISS International Partner JAXA; the World Biomimetic Foundation in Spain who is interested in advancing the use of biomimicry to provide technical solutions in many industries; Satellite Application Catapult in London, England who interested in pursuing U.S. collaborations with the Space and Life Sciences Innovation Centre under development in Scotland; and DLR in Cologne, Germany who developed :envihab, a collaborative facility for partners to pursue research and technology projects of mutual interest. The NHHPC has sponsored two global networking forums on innovation by partners Wyle, NASA, and DLR, was featured in the 2013 Humans in Space Symposium Panel on "NHHPC and :envihab - reach out to Future Markets," and is working on an international meeting for Spring 2014 in Cologne with :envihab.

Davis, Jeffrey R.↗
DOE OSTI2025

femto-PIXAR: a self-supervised neural network method for reconstructing femtosecond X-ray free electron laser pulses

X-ray Free Electron Lasers (X-FELs) operate in a wide range of lasing configurations for a broad variety of scientific applications at ultrafast time-scales such as structural biology, materials science, and atomic and molecular physics. Shot-by-shot characterization of the X-FEL pulses is crucial for analysis of many experiments as well as tuning the X-FEL performance. However, for the weak pulses found in advanced configurations, e.g. those needed for coherent, two-pulse studies of quantum materials, there is no current method for reliably resolving pulse profiles. Here we show that a physics-based U-net model can reconstruct the individual pulse power profiles for sub-picosecond pulse separation without the need for simulations. Using experimental data from weak X-FEL pulse pairs, we demonstrate we can learn the pulse characteristics on a shot-by-shot basis when conventional methods fail.

43 PARTICLE ACCELERATORS↗
DOE OSTI2026

In-situ sensor monitoring of multi-class gas porosity formation in laser powder bed fusion using convolutional neural network

In-situ monitoring of defect formation remains a significant challenge in the laser powder bed fusion (LPBF) process. Recent advances have enabled real-time defect detection with machine learning and in-situ sensing technologies; however, most studies focus on binary classification of keyhole pores, limiting nuanced multi-class pore differentiation and formation mechanisms. This work introduces a multi-class pore detection framework (no pore, small pores < 15 µm, and large pores > 15 µm) by leveraging photodiode sensor data alongside high-fidelity synchrotron X-ray imaging. The 15 µm threshold is selected to distinguish between two fundamentally different defect mechanisms, following the physical size-mechanism boundary established by prior high-resolution synchrotron X-ray characterization of Al6061 LPBF. Distinguishing these classes is critical because large keyhole pores are structurally detrimental, whereas small gas pores are often benign, requiring different process control strategies. Thermal emission monitoring data collected simultaneously with high-speed X-ray imaging at the Stanford Synchrotron Radiation Lightsource (SSRL), are correlated with subsurface melt pool dynamics to establish ground truth. Continuous Wavelet Transform (CWT) with optimized parameters converts the photodiode time-series signals into time–frequency images, facilitating feature extraction. Convolutional Neural Networks (CNN) are then applied for real-time multi-class pore classification in an average inference time of 1 ms per signal window. It achieves 79% accuracy and an Area Under the Receiver Operating Characteristic curve (AUC ROC) score of 0.89 with five-fold cross-validation. The results demonstrate that coupling CWT-based feature engineering with CNN architecture enables reliable multi-class pore detection in Al6061 builds using affordable in-situ sensors. This approach advances scalable and affordable quality assurance in additive manufacturing by moving beyond binary defect detection toward more nuanced classification of porosity mechanisms with in-situ sensors and machine learning.

Laser powder bed fusion, Multi-class pores, In-sit↗
NASA NTRS2013

SpF: Enabling Petascale Performance for Pseudospectral Dynamo Models

Pseudospectral (PS) methods possess a number of characteristics (e.g., efficiency, accuracy, natural boundary conditions) that are extremely desirable for dynamo models. Unfortunately, dynamo models based upon PS methods face a number of daunting challenges, which include exposing additional parallelism, leveraging hardware accelerators, exploiting hybrid parallelism, and improving the scalability of global memory transposes. Although these issues are a concern for most models, solutions for PS methods tend to require far more pervasive changes to underlying data and control structures. Further, improvements in performance in one model are difficult to transfer to other models, resulting in significant duplication of effort across the research community.We have developed an extensible software framework for pseudospectral methods called SpF that is intended to enable extreme scalability and optimal performance. High-level abstractions provided by SpF unburden applications of the responsibility of managing domain decomposition and load balance while reducing the changes in code required to adapt to new computing architectures. The key design concept in SpF is that each phase of the numerical calculation is partitioned into disjoint numerical kernels that can be performed entirely in-processor. The granularity of domain-decomposition provided by SpF is only constrained by the data-locality requirements of these kernels. SpF builds on top of optimized vendor libraries for common numerical operations such as transforms, matrix solvers, etc., but can also be configured to use open source alternatives for portability. SpF includes several alternative schemes for global data redistribution and is expected to serve as an ideal testbed for further research into optimal approaches for different network architectures.In this presentation, we will describe the basic architecture of SpF as well as preliminary performance data and experience with adapting legacy dynamo codes. We will conclude with a discussion of planned extensions to SpF that will provide pseudospectral applications with additional flexibility with regard to time integration, linear solvers, and discretization in the radial direction.

Pseudospectral (PS)↗
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

HydraGNN_Predictive_GFM_2026 - Ensemble of predictive graph foundation models for atomistic materials modeling

This release contains data and parameters of HydraGNN-based graph foundation models trained as a result of the work published in the pre-print "Exascale Multi-Task Graph Foundation Models for Imbalanced, Multi-Fidelity Atomistic Data" by M. Lupo Pasini et al. (https://arxiv.org/abs/2604.15380). We jointly train on 16 open first-principles datasets (544+ million structures covering 85+ elements) using a multi-task architecture with per-dataset heads and a scalable ADIOS2/DDStore data pipeline. On Frontier, we execute six large-scale DeepHyper hyperparameter optimization campaigns in FP64 and promote the top-performing message-passing models to sustained 2,048-node training, yielding a PaiNN-based lead model. The version of HydraGNN used to generate the outputs provided in this release is HydraGNN v5.0 (https://github.com/ORNL/HydraGNN/releases/tag/v5.0) The list of datasets used for the training of the graph foundation model is the following: 1) Alexandria [1] 2) ANI1x [2] 3) MPTrj [3] 4) Open Catalyst 2020 (OC20) [4] 5) Open Catalyst 2022 (OC22) [5] 6) Open Catalyst 2025 (OC25) [6] 7) Open Direct ir Capture 2023 (ODAC23) [7] 8) Open Materials 2024 (OMat24) [8] 9) Open Molecules 2025 (OMol25) [9] 10) OMol25-neutral (subset of OMol25 that contains only molecules with zero total charge) 11) OMol25-non-neutral (subset of OMol25 that contains only molecules with non-zero total charge) 12) Open Polymers 2026 (OPoly2026) [10] 13) Nabla2DFT [11] 14) QCML [12] 15) QM7X [reference 13] 16) transition1x [14] Dataset references: [1] J. Schmidt et al., “A dataset of 175k stable and metastable materials calculated with the PBEsol and SCAN functionals,” Scientific Data, vol. 9, p. 64, 2022. [2] J. S. Smith et al., “The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules,” Scientific Data, vol. 7, p. 134, 2020. [Online]. Available: https: //www.nature.com/articles/s41597-020-0473-z [3] A. Jain et al., “Commentary: The Materials Project: A materials genome approach to accelerating materials innovation,” APL Materials, vol. 1, no. 1, p. 011002, 07 2013. [Online]. Available: https://doi.org/10.1063/1.4812323 [4] L. Chanussot et al., “Open catalyst 2020 (oc20) dataset and community challenges,” ACS Catalysis, vol. 11, no. 10, pp. 6059–6072, 2021. [Online]. Available: https://doi.org/10.1021/acscatal.0c04525 [5] K. Tran et al., “Open catalyst 2022 (oc22) dataset and challenges for oxidation electrocatalysts,” ACS Catalysis, vol. 13, no. 5, pp. 3066–3084, 2023. [Online]. Available: https://doi.org/10.1021/acscatal.2c05426 [6] S. J. Sahoo et al., “The open catalyst 2025 (oc25) dataset and models for solid-liquid interfaces,” arXiv preprint arXiv:2509.17862, 2025. [Online]. Available: https://arxiv.org/abs/2509.17862 [7] A. Sriram et al., “The open DAC 2023 dataset and challenges for sorbent discovery in direct air capture,” ACS Central Science, vol. 10, no. 5, pp. 923–941, 2024. [8] L. Barroso-Luque et al., “Open materials 2024 (omat24) inorganic materials dataset and models,” 2024. [Online]. Available: https://arxiv.org/abs/2410.12771 [9] D. S. Levine et al., “The open molecules 2025 (OMol25) dataset, evaluations, and models,” 2025. [Online]. Available: https://arxiv.org/abs/2505.08762 [10] D. S. Levine et al., The open polymers 2026 (OPoly26) dataset and evaluations,” arXiv preprint arXiv:2512.23117, 2025. [Online]. Available: https://arxiv.org/abs/2512.23117 [11] K. Khrabrov et al., “Nabla2dft: A universal quantum chemistry dataset of drug-like molecules and a benchmark for neural network potentials,” in NeurIPS 2024 Datasets and Benchmarks Track, 2024. [Online]. Available: https://openreview.net/forum?id=ElUrNM9U8c [12] S. Ganscha et al., “The QCML dataset, quantum chemistry reference data from 33.5M DFT and 14.7B semi-empirical calculations,” Scientific Data, vol. 12, p. 406, 2025. [13] J. Hoja et al., “QM7-X, a comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules,” Scientific Data, vol. 8, p. 43, 2021. [Online]. Available: https://www.nature.com/articles/s41597-021-00812-2 [14] M. Schreiner et al., “Transition1x - a dataset for building generalizable reactive machine learning potentials,” Scientific Data, vol. 9, p. 779, 2022. The folder "datasets_ADIOS2_format" contains the set of pre-processed datasets in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used for the development and training of GFMs in this work. The "datasets_ADIOS2_format" directory contains 2 sub-directories, one for the version "v1" of the datasets and one for the version "v2" of the datasets. The version "v1" of the datasets provides values of the total energy as they are extracted from the original data as it was released by the respective institutions. The version "v2" of the datasets provides values of the energy that have been realigned. The realignment was performed by training a linear regression model that predicts the total energy as a function of the chemical composition of the atomistic structure, and then subtract such prediction from the original value of the total energy. Both folders "v1" and "v2" contain 16 sub-directories, each corresponding to an ADIOS2-formatted dataset The folder "DeepHyper-results" contains the configurational files and model's parameters for all the 186 HPO trials that were successfully completed by the scalable hyperparameter optimization (HPO) runs on Frontier. The content of the folder "DeepHyper-results" I structured as follows: 1) task-list.txt: list of mpnn name, jobid, and deephyper task id 2) gfm_${MPNN}_${JOBID}_0.${TASKID}: run directory with checkpoint files 3) gfm_${MPNN}: deephyper summary directory (*.csv) for each specific MPNN type 4) deephyper-experiment-${JOBID}: output and error logs for each job The file "deephyper-sorted.csv" contains the details of each HydraGNN model built and tested by HPO, obtained by merging the (*.csv) filed from each HPO run executed. Out of all the HPO trials, we selected 10 to continue the training of the respective HydraGNN models. Due to limited computational budget available in the LRN070 allocation we could not complete the training till convergence for all these 10 selected models. The folder "models" contains multiple sub-folders, one per each HydraGNN model trained. Each model sub-folder contains the parameters of each HydraGNN model, with multiple checkpoint-restarts. The list of sub-folders are as follows: 1) multidataset_hpo-BEST1-fp64 2) multidataset_hpo-BEST2-fp64 3) multidataset_hpo-BEST3-fp64 4) multidataset_hpo-BEST4-fp64 5) multidataset_hpo-BEST5-fp64 6) multidataset_hpo-BEST6-fp64 7) multidataset_hpo-BEST7-fp64 8) multidataset_hpo-BEST8-fp64 9) multidataset_hpo-BEST9-fp64 10) multidataset_hpo-BEST10-fp64 Within each one of these folders, additional auxiliary log files are provided with descriptions about how the training proceeded. The lead PaiNN-model is contained inside "multidataset_hpo-BEST6-fp64". The file "mlp_branch_weights" contains the parameters of the multi-layer perceptron (MLP) used to reconcile the predictions of the 16 output decoding heads of the HydragNN architectures. The MLP takes in input the chemical composition of the atomistic structure and predicts averaging weights to linearly mix the predictions of each output decoding head toward consolidating them into a single one. The folder "1.1billion-structure-inference" contains 1.1 billion atomistic structures randomly generated. Each structures is associated with energy and forces predicted with the lead-PaiNN model combined with the MLP model for reconciliation of the multi-branch predictions generated by the 16 output decoding heads. The folder "1.1billion-structure-inference" contains 9,300 (*.tar.gz) subdirectories, one per Frontier compute node used to execute the inference at exascale. Once uncompressed, each (*.tar.gz) subdirectory contains an ADIOS2 (*.bp) file container, where each atomistic structure is stored as a PyTorch-Geometric Data object. The file "export_dataset_environment_variables.sh" contains the environment variables that need to be set before running the HydraGNN code to reproduce the results provided in this dataset release. The code that can be used to load the ADIOS2 files, load HydraGNN models, and run inference is available at: https://github.com/ORNL/HydraGNN/releases/tag/v5.0

36 MATERIALS SCIENCE↗
DOE OSTI2024

Predicting Missing Regions in Charged Particle Tracks Using a Sparse 3D Convolutional Neural Network

The 2x2 Demonstrator is a prototype detector for the Deep Underground Neutrino Experiment (DUNE)'s Near Detector. Both the 2x2 Demonstrator and the Near Detector itself will have inactive regions wherein there is no sensitivity to charge deposition and light signals that arise from charged particle interactions with liquid argon. In the 2x2, these inactive regions are positioned in-between the active detector modules, which introduces the challenge of inferring what charge signals ought to look like in these regions. This study explores the use of a Sparse 3D Convolutional Neural Network (ConvNet) to infer missing regions in charged particle tracks. Hits corresponding to energy depositions are voxelized into a three-dimensional (3D) grid for each track. Inactive regions within the tracks are replaced with a dense, rectangular 3D grid of voxels, ensuring consistent step sizes in X, Y, and Z directions. Voxels in these dense regions are initialized with an energy value of -1, indicating nonphysical energy or charge. The model is trained to predict which voxels should activate as part of the track and which should not, with the goal of eventually inferring the missing charge or energy values in these voxels. Results indicate that the model accurately predicts track voxels within ±1 unit in X, Y, or Z directions and effectively identifies non-track voxels, despite some overprediction. The approach shows promise in prediction of missing track regions with some accuracy.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Resolving three-dimensional nanoscale heterogeneities in lithium metal batteries with cryoelectron tomography

Current direct observation of sensitive battery materials and interfaces primarily relies on two-dimensional (2D) imaging, leaving out their three-dimensional (3D) relationship. Here, in this study, we used cryoelectron tomography (cryo-ET) to visualize the lithium metal anode in 3D at nanometer resolution and cryoelectron microscopy (cryo-EM) to reveal atomic details in local regions. We imaged both freshly prepared and calendar-aged Li metal anodes to reveal the development of LiH in Li dendrites and the Li-LiH interface, as well as the development of the solid-electrolyte interphase (SEI). Using a convolutional neural network-based technique, the 3D arrangement of Li metal, along with nanoscale LiH and Cu heterogeneities in dendrites, was visualized and annotated. In longer-term calendar aging, we observed more substantial LiH growth accompanied by extended SEI growth. Our results show that the growth of LiH and the extended SEI during battery calendar aging are temporally and spatially separate processes.

LiH↗
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 ↗