Results for “large-scale inference”

DOE OSTI2026

Data-driven particle dynamics: Structure-preserving coarse-graining for emergent behavior in non-equilibrium systems

Multiscale systems are ubiquitous in science and technology, but are notoriously challenging to simulate as short spatiotemporal scales must be appropriately linked to emergent bulk physics. When expensive high-dimensional dynamical systems are coarse-grained into low-dimensional models, the entropic loss of information leads to emergent physics which are dissipative, history-dependent, and stochastic. To machine learn coarse-grained dynamics from time-series observations of particle trajectories, we propose a framework using the metriplectic bracket formalism that preserves these properties by construction; most notably, the framework guarantees discrete notions of the first and second laws of thermodynamics, conservation of momentum, and a discrete fluctuation-dissipation balance crucial for capturing non-equilibrium statistics. We introduce the mathematical framework abstractly before specializing to a particle discretization. As labels are generally unavailable for entropic state variables, we introduce a novel self-supervised learning strategy to identify emergent structural variables. We validate the method on benchmark systems and demonstrate its utility on two challenging examples: (1) coarse-graining star polymers at challenging levels of coarse-graining while preserving non-equilibrium statistics, and (2) learning models from high-speed video of colloidal suspensions that capture coupling between local rearrangement events and emergent stochastic dynamics. We provide open-source implementations in both PyTorch and LAMMPS, enabling large-scale inference and extensibility to diverse particle-based systems.

Computational Engineering, Finance, and Science (c↗

DOE OSTI2026

Probing jet base emission of M87* with the 2021 Event Horizon Telescope observations

We investigate the presence and spatial characteristics of the jet base emission in M87* at 230 GHz, enabled by the significantly enhanced (u,v) coverage in the 2021 Event Horizon Telescope (EHT) observations. The integration of the 12−m Kitt Peak Telescope (USA) and NOEMA (France) stations into the array introduces two critical intermediate-length baselines to SMT (USA) and IRAM 30−m (Spain), providing sensitivity to emission structures at spatial scales of ∼250 μas and ∼2500 μas (∼ 0.02 pc and ∼ 0.02 pc). Without these new baselines, previous EHT observations of the source in 2017 and 2018 lacked the capability to constrain emission on large scales, where a “missing flux” of order ∼1 Jy is expected to reside. To probe these scales, we analyzed closure phases–robust against station-based gain calibration errors–and model the jet base emission using a simple Gaussian component offset from the compact ring emission at spatial separations > 100 μas. Our analysis revealed a Gaussian feature centered at (ΔRA ≈ 320 μ as, ΔDec. ≈ 60 μ as), projected separation of ≈ 5500 AU, with an estimated flux density of only ∼60 mJy, implying that most of the missing flux identified in previous EHT studies had to originate from different, larger scales. Brighter emission at the relevant spatial scales is firmly ruled out, and the data do not favor more complex models. This component aligns with the inferred position of the large-scale jet and is therefore physically consistent with the emission of the jet base. While our findings point to detectable jet base emission at 230 GHz, the limited coverage provided by only two intermediate baselines limits our ability to robustly reconstruct its morphology. Consequently, we treated the recovered Gaussian as an upper limit on the jet base flux density. Future EHT observations with expanded intermediate baseline coverage will be essential to constrain the structure and nature of this component with higher precision.

accretion, accretion disks↗

DOE OSTI2026

E-PINNs: Epistemic Physics-Informed Neural Networks

Physics-informed neural networks (PINNs) have demonstrated promise as a framework for solving forward and inverse problems involving partial differential equations. Despite recent progress in the field, it remains challenging to quantify uncertainty in these networks. While techniques such as Bayesian PINNs (B-PINNs) provide a principled approach to capturing epistemic uncertainty through Bayesian inference, they can be computationally expensive for large-scale applications. In this work, we propose Epistemic Physics-Informed Neural Networks (E-PINNs), a framework that uses a small network, the epinet, to efficiently quantify epistemic uncertainty in PINNs. The proposed approach works as an add-on to existing, pre-trained PINNs with a small computational overhead. We demonstrate the applicability of the proposed framework in various test cases and compare the results with B-PINNs using Hamiltonian Monte Carlo (HMC) posterior estimation and dropout-equipped PINNs (Dropout-PINNs). In our experiments, E-PINNs achieve calibrated coverage with competitive sharpness at substantially lower cost. We demonstrate that when B-PINNs produce narrower bands, they under-cover in our tests. E-PINNs also show better calibration than Dropout-PINNs in these examples, indicating a favorable accuracy-efficiency trade-off.

AI for Science↗

DOE OSTI2026

Dark Energy Survey Year 3 results: Simulation-based 𝑤CDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design

Data-driven approaches using deep learning are emerging as powerful techniques to extract non-Gaussian information from cosmological large-scale structure. Here, this work presents the first simulation-based inference (SBI) pipeline that combines weak lensing and galaxy clustering maps in a realistic Dark Energy Survey Year 3 (DES Y3) configuration and serves as preparation for a forthcoming analysis of the survey data. We develop a scalable forward model based on the CosmoGridV1 suite of N-body simulations to generate over one million self-consistent mock realizations of DES Y3 at the map level. Leveraging this large dataset, we train deep graph convolutional neural networks on the full survey footprint in spherical geometry to learn low-dimensional features that approximately maximize mutual information with target parameters. These learned compressions enable neural density estimation of the implicit likelihood via normalizing flows in a ten-dimensional parameter space spanning cosmological 𝑤CDM, intrinsic alignment, and linear galaxy bias parameters, while marginalizing over baryonic, photometric redshift, and shear bias nuisances. To ensure robustness, we extensively validate our inference pipeline using synthetic observations derived from both systematic contaminations in our forward model and independent Buzzard galaxy catalogs. Our forecasts yield significant improvements in cosmological parameter constraints, achieving 2−3× higher figures of merit in the 𝛺 𝑚 − 𝑆 8 plane relative to our implementation of baseline two-point statistics and effectively breaking parameter degeneracies through probe combination. These results demonstrate the potential of SBI analyses powered by deep learning for upcoming Stage-IV wide-field imaging surveys.

Thomsen, A. [Zurich, ETH] (ORCID:0000000203099021)↗

DOE OSTI2026

Tomography of the gamma-ray sky from cross-correlation with DESI DR2 and unWISE galaxies

We study the origin of extragalactic gamma-ray emission observed by Fermi-LAT, using the cross-correlation of the gamma-ray sky with maps of large-scale structure provided by the DESI and unWISE surveys. Tomographic cross-correlation reveals the bias-weighted redshift distributions of gamma-ray sources. We first illustrate this method by cross-correlating detected gamma-ray point sources with large-scale structure. We find a significant cross-correlation and infer a point source redshift distribution broadly consistent with the distribution of identified optical counterparts previously reported in the literature, as well as a similar linear bias ($b \approx 2$) to massive galaxies that host bright active galactic nuclei. We then study the clustering of the Fermi unresolved gamma-ray background (UGRB), both in auto-correlation and in cross-correlation with large-scale structure. We detect the cross-correlation of the UGRB and LSS at $\sim 10σ$ in total, with highly significant detections from both DESI and unWISE. Our measurements suggest that the redshift distribution of the UGRB is broadly consistent with the redshift distribution of detected point sources. Additionally, we find a relatively weak amplitude for the cross-correlation with large-scale structure at z < 2, suggesting a significant fraction of the UGRB does not come from z < 2 large-scale structure. A natural candidate is contamination of from residual Galactic emission, and our best estimate of the contamination level derived from the UGRB auto-spectrum suggests that the mean bias of UGRB sources is indeed quite similar to the bias of detected Fermi point sources. However, we cannot exclude additional emission from gamma-ray sources at high redshift, z > 2, and we suggest that cross-correlation with tracers at z > 2, including CMB lensing, would be the ideal way to determine the fraction of z > 2 emission.

Krolewski, Alex [Waterloo U., Math. Dept.; Waterlo↗

DOE OSTI2026

Inference as a Service for HEP & NP experiments

This work presents how American Science Cloud resources and services can accelerate scientific discovery across large-scale HEP & NP experiments. Our demonstrators include use cases from intensity and energy frontiers. This is a first end-to-end multi-experiment, multi-facility demonstration of AmSC platforms.

Bhattacharya, M. [Fermilab]↗

DOE OSTI2026

GP Cosmology Surrogate v1.0

GP Cosmology Surrogate is a Python library for building and training a generalized multi-output Gaussian process (GP) framework of @takhtaganov2021cosmic. In this approach, the surrogate is constructed sequentially, guided by a Bayesian optimization acquisition function that targets reduction of emulation error in the regions most consistent with the observational data. This adaptive design concentrates computational resources where they have the greatest impact on inference accuracy. The library supports efficient training for separable GP kernels, which allows the use of Kronecker algebra to handle high-dimensional input spaces and large numbers of correlated outputs. This makes it well suited for applications such as modeling cosmological power spectra, large-scale physical simulations, and multi-output hyperparameter tuning. By combining scalable multi-output GP modeling with data-driven adaptive sampling, GPsurrogate enables parameter inference and optimization with substantially fewer simulations than conventional space-filling designs.

Lukic, Zarija [Lawrence Berkeley National Laborato↗

DOE OSTI2026

From IMT Device Measurements to Network-Level Consequences: When Learning Suppresses Beyond-LIF Neuron Dynamics

Emerging neuromorphic devices such as insulator--metal transition (IMT) devices exhibit complex temporal dynamics, including slow internal state memory, hysteresis, and burst-like firing, which are poorly captured by conventional leaky integrate-and-fire (LIF) neurons. However, it remains unclear when such dynamics influence learning and inference at the network level, particularly under commonly used unsupervised plasticity rules. We present a controlled, full-stack co-design study spanning experimental characterization of individual IMT devices, compact neuron model development, and large-scale spiking network simulations with identical architectures and learning rules. Rather than optimizing benchmark accuracy, our goal is to diagnose when neuron-level dynamics survive learning and competition, and when they are suppressed, to inform the co-design of devices, networks, and learning rules that can exploit beyond-LIF complexity.

42 ENGINEERING↗

DOE OSTI2026

Multiprobe constraints on early and late time dark energy

We perform a multiprobe analysis combining cosmic microwave background (CMB) data from Planck and the Atacama Cosmology Telescope (ACT), ACT CMB lensing, and large-scale structure (LSS) measurements from the Dark Energy Spectroscopic Instrument (DESI), including DESI Legacy Imaging Survey (LS) galaxies and baryon acoustic oscillations (BAOs). We present the first 5×2 pt analysis of ACT DR6 lensing, DESI LS, and Planck Integrated Sachs-Wolfe (ISW). Within Λ⁢CDM, this yields 𝑆 8 =𝜎 8 ⁢(Ω 𝑚 /0.3) 0.5 =0.819 ±0.016, in good agreement with primary CMB inferences and provides a sound-horizon-free Hubble constant constraint of 𝐻 0 =70.0±4.4 km s −1 Mpc −1 . Then, combining with CMB primary and BAO, we reconfirm a CMB–BAO discrepancy in the Ω 𝑚 –$\frac{𝐷𝑣}{𝑟𝑑}$ plane, which is heightened when combining BAO with the 5 ×2 pt data vector. We explore two dark-energy extensions that may reconcile this: an early-time modification, early dark energy (EDE), and late-time dynamical dark energy (DDE) parametrized by 𝑤 0 ⁢𝑤 𝑎 . For CMB primary +BAO +5 ×2 pt, we find a 3.3⁢𝜎 preference for DDE over Λ⁢CDM, while EDE is modestly favored at 2.3⁢𝜎. The models address different shortcomings of Λ⁢CDM: DDE relaxes the neutrino mass bound (𝑀 𝜈 <0.17 eV vs <0.050 eV under Λ⁢CDM), making it compatible with neutrino oscillation measurements, while EDE raises the Hubble constant to 𝐻 0 =70.5±1.2 km s −1 Mpc −1 , easing the discrepancy with SH0ES. However, neither model resolves both issues simultaneously. Our analysis indicates that both DDE and EDE remain viable extensions of Λ⁢CDM within current uncertainties and demonstrates the capacity of combined probes to place increasingly stringent constraints on cosmological parameters.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2026

Microsecond-latency feedback at a particle accelerator by online reinforcement learning on hardware

The commissioning and operation of future large-scale scientific experiments will challenge current tuning and control methods. Reinforcement learning (RL) algorithms are a promising solution due to their ability to dynamically adapt to changing environments and consider delayed consequences. In many real-world applications, RL policies must produce actions in real time, often within microseconds to milliseconds, imposing significant constraints on system latency and computational overhead that conventional machine learning libraries are not designed to handle. To control phenomena in real time at these timescales, RL needs to be deployed on-the-edge, namely on dedicated hardware located near the system it controls, without relying on a host CPU or cloud-based inference. In this work we present the design and deployment of an experience accumulator system in a particle accelerator. In this system, deep-RL algorithms run using hardware acceleration and act within a few microseconds, enabling the use of RL for control of phenomena like beam instabilities. The training uses the collected data offline to reduce the number of operations carried out on the acceleration hardware. The proposed architecture was tested in real experimental conditions at the Karlsruhe research accelerator, a synchrotron light source, where the system was used to control artificially induced horizontal betatron oscillations in real-time, with a control loop period of just 2.7 μs. The results showed a performance comparable to the commercial feedback system available at the accelerator, demonstrating the viability and potential of this approach. Due to the self-learning and reconfiguration capability of this implementation, a seamless application to other control problems is possible. Applications range from particle accelerators to large-scale research and industrial facilities.

FPGA↗

DOE OSTI2026

Dataset of Generative AI Workload Power Profiles

This dataset provides a collection of high-resolution (5/10 Hz or every 0.2/0.1 seconds) power consumption profiles for generative artificial intelligence (GenAI) workloads executed on NLR's High Performance Computing (HPC) platform Kestrel. The dataset also includes examples of representative whole-facility power profiles generated using a bottom-up, event-driven, data center energy model . This dataset is designed to support research in energy modeling, infrastructure planning, energy system integration, and sustainability analysis for AI-driven computing systems. The dataset captures time-resolved electrical power measurements across a diverse set of configurations, including variations in job type (inference vs. training), workload (LLM vs. image generation), datasets, and number of compute nodes. Power traces are provided in a standardized format and include both raw/instantaneous and aggregated files. Each profile is accompanied by metadata describing workload parameters, enabling reproducibility and cross-study comparison. The dataset is intended for use in applications such as data center infrastructure planning, energy modeling, demand response and grid impact studies, and development and validation of system-level simulation tools. By making these workload-specific power profiles publicly available, this dataset aims to address the current lack of open, empirical energy data for generative AI systems and to facilitate transparent, reproducible research on the energy and environmental impacts of large-scale AI deployment. If you use this dataset, please cite the associated publication: Vercellino et al., “Measurement of Generative AI Workload Power Profiles for Whole-Facility Data Center Infrastructure Planning,” arXiv:2604.07345 (2026).

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

Learning genetic perturbation effects with variational causal inference

Advances in sequencing technologies have enhanced the understanding of gene regulation in cells. In particular, Perturb-seq has enabled high-resolution profiling of the transcriptomic response to genetic perturbations at the single-cell level. This understanding has implications in functional genomics and potentially for identifying therapeutic targets. Various computational models have been developed to predict perturbational effects. While deep learning models excel at interpolating observed perturbational data, they tend to overfit in the lack of enough data and may not generalize well to unseen perturbations. In contrast, mechanistic models, such as linear causal models based on gene regulatory networks, hold greater potential for extrapolation, as they encapsulate regulatory information that can predict responses to unseen perturbations. However, their application has been limited to small studies due to overly simplistic assumptions, making them less effective in handling noisy, large-scale single-cell data. We propose a hybrid approach that combines a mechanistic causal model with variational deep learning, termed Single Cell Causal Variational Autoencoder (SCCVAE). The mechanistic model employs a learned regulatory network to represent perturbational changes as shift interventions that propagate through the learned network. SCCVAE integrates this mechanistic causal model into a variational autoencoder, generating rich, comprehensive transcriptomic responses. Our results indicate that SCCVAE exhibits superior performance over current state-of-the-art baselines for extrapolating to predict unseen perturbational responses. Additionally, for the observed perturbations, the latent space learned by SCCVAE allows for the identification of functional perturbation modules and simulation of single-gene knockdown experiments of varying penetrance, presenting a robust tool for interpreting and interpolating perturbational responses at the single-cell level.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

Probing cosmic velocities with the pairwise kinematic Sunyaev-Zel’dovich signal in DESI Bright Galaxy Sample DR1 and ACT DR6

We present a measurement of the pairwise kinematic Sunyaev-Zel’dovich (kSZ) signal using the Dark Energy Spectroscopic Instrument (DESI) Bright Galaxy Sample (BGS) Data Release 1 (DR1) galaxy sample overlapping with the Atacama Cosmology Telescope (ACT) CMB temperature map. Our analysis makes use of 1.6 million galaxies with stellar masses log⁡ 𝑀 ⋆ /𝑀 ⊙ >10, and we explore measurements across a range of aperture sizes (2.1′ <𝜃 ap <3.5′) and stellar mass selections. This statistic directly probes the velocity field of the large-scale structure, a unique observable of cosmic dynamics and modified gravity. In particular, at low redshifts, this quantity is especially interesting, as deviations from General Relativity are expected to be largest. Notably, our result represents the highest-significance low-redshift (𝑧 ∼ 0.3) detection of the kSZ pairwise effect yet. In our most optimal configuration (𝜃 ap =3.3′, log⁡ 𝑀 ⋆ >11), we achieve a 5⁢𝜎 detection. Assuming that an estimate of the optical depth and galaxy bias of the sample exists via e.g., external observables, this measurement constrains the fundamental cosmological combination 𝐻 0 ⁡𝑓⁡𝜎$^2_8$. A key challenge is the degeneracy with the galaxy optical depth. We address this by combining CMB lensing, which allows us to infer the halo mass and galaxy population properties, with hydrodynamical simulation estimates of the mean optical depth, $\bar{𝜏}$ . We stress that this is a proof-of-concept analysis; with BGS DR2 data we expect to improve the statistical precision by roughly a factor of two, paving the way toward robust tests of modified gravity with kSZ-informed velocity-field measurements at low redshift.

Hadzhiyska, Boryana [Institute of Astronomy; Kavli↗

DOE OSTI2026

Assessing Metal Ion Assignment Accuracy in Protein Data Bank Models via Elemental Spectroscopy

Accurate representation of metal ions in macromolecular structures is critical for chemical interpretation, computational modeling, and machine-learning methods that rely on Protein Data Bank (PDB) entries. However, the elemental identity of metals modeled in crystallographic structures is often inferred indirectly and rarely validated experimentally. Here, we combine Particle Induced X-ray Emission (PIXE) and X-ray Fluorescence Spectroscopy (XRFS) to determine the elemental composition of protein samples used to generate 70 deposited metalloprotein crystal structures. By analyzing the original protein material employed for crystallization, but before the addition of crystallization buffer solutions, we assess whether the modeled metal ions in deposited structures are consistent with experimentally detectable elemental content. We find that in a majority of cases, the metals modeled in the corresponding PDB entries are inconsistent with the metals present in the protein samples before crystallization, or that additional metals are present but not represented in the structural models. Spectroscopic results were integrated with automated crystallographic validation metrics, including real-space Z-difference (RSZD) analysis and systematic rerefinement, to evaluate atomic-number mismatch at metal sites. PIXE and XRFS show strong agreement for dominant elemental signals and provide complementary, scalable approaches for identifying suspect metal assignments. This work does not address physiological or functional metalation but instead highlights a widespread data integrity issue in deposited macromolecular structures, PDB-wide. These results establish an experimentally corroborated link between elemental identity and crystallographic validation metrics, enabling the large-scale detection of chemically inconsistent annotations in structural databases used for computational modeling and machine learning.

Crystallization↗

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 OSTI2026

Seismic Elastic Double-Beam Characterization of Faults and Fractures for CO₂ Storage Site Selection

Site characterization for underground injection and storage of gigatonne-scale CO₂ requires reliable and cost-effective methods to detect and characterize faults and fractures and to assess their stress state and fault activation potential. This is critical, as wastewater injection and disposal have been shown to activate faults and induce earthquakes, and CO₂ leakage remains a key concern for long-term storage. In this project, we developed seismic methods to detect and characterize large-scale sedimentary and crystalline basement faults and associated small-scale fractures below conventional seismic imaging resolution using multicomponent (9C) surface seismic data. Machine learning was used to automatically interpret large-scale faults, providing key information for estimating the maximum magnitude of potential induced earthquakes. High-fidelity imaging was achieved by exploiting redundancy across multiple elastic wave modes, where independent images from different modes and frequencies cross-validate each other. We also used our nonlinear signal comparison (NLSC) method for ground roll removal, improving data quality in complex near-surface conditions. The methods were validated using field data acquired in central Montana. Results show that basement faults extend into the sedimentary section and that small-scale fractures are widespread above the basement. The inferred stress orientation is consistent with regional stress data, and the estimated maximum induced earthquake magnitude is small (Mw ~2.3). The developed workflow provides a practical approach for fault and fracture characterization and for assessing induced seismicity and leakage risk. It is directly applicable to CO₂ storage site selection and to other subsurface systems.

02 PETROLEUM↗

NASA NTRSDate not supplied

ARCTRON: A Rapid Experimental Proving Ground for TPS Experiments and Arcjet Technology Development

Innovation in high-enthalpy facilities is fundamentally limited by the cost and risk of experimentation. New concepts for plasma control, diagnostics, facility components, and plasma-material interaction often require repeated iterations that are impractical to perform in production arcjets. As a result, promising ideas may remain unexplored or reach operational facilities only after significant development effort. ARCTRON is being developed as a rapid experimental proving ground where new ideas in plasma science, arcjet engineering, diagnostics, and material response can be conceived, tested, and quantitatively evaluated before transition to large-scale facilities. The platform combines radio-frequency (RF) and DC arc plasma generation, externally applied magnetic fields, configurable gas composition, reduced-pressure operation, laser heating, electrical biasing, and modular diagnostic access. These capabilities permit the plasma source, applied forcing, test article, and measurement configuration to be modified independently, allowing individual physical mechanisms to be isolated more readily than in a traditional test environment. One class of investigations addresses fundamental plasma-surface interaction physics. Conventional material tests often expose a specimen simultaneously to convective heating, reactive species, pressure, shear, radiation, and surface-current effects. The resulting material response may be measured accurately, while the contribution of each mechanism remains difficult to identify. ARCTRON is designed to vary these effects selectively. Plasma chemistry can be changed independently through configurable gas mixtures; magnetic fields and electrical biasing can modify charged-particle transport; laser heating can provide a non-plasma thermal input; and pressure, flow, and discharge mode can be varied over a broad operating space. This enables controlled tests of hypotheses involving surface catalycity, reactive-species transport, plasma-assisted oxidation, electromagnetic effects, shear, and the relative contributions of thermal and chemical loading. A second class of investigations enabled by this approach concerns the engineering of high-enthalpy facilities themselves. Arc-heated facilities are limited by electrode erosion, unstable arc attachment, localized heating, and damage to nozzles and other plasma-facing components. ARCTRON provides a lower-cost environment for testing concepts intended to mitigate these limitations. Candidate investigations include the use of applied magnetic fields to alter current paths and reduce plasma interaction with nozzle walls, ExB forcing to introduce controlled plasma rotation, magnetic or geometric approaches for distributing arc attachment, and alternative electrode or discharge configurations intended to reduce erosion and improve stability. Because the platform is reconfigurable, these concepts can be evaluated through repeated design--build--test cycles before they are considered for implementation in operational facilities. The platform also supports the development and validation of diagnostics that may be difficult to introduce initially into a large arcjet. Current and planned measurements include spatially resolved optical emission spectroscopy, electrostatic probes, fast imaging, pyrometry, calorimetry, laser-induced fluorescence, and absorption spectroscopy. These diagnostics are intended not merely to document a nominal operating condition, but to constrain the local plasma state and its relationship to component or material response. The modular facility geometry allows diagnostic concepts to be tested, calibrated, and compared under repeatable conditions before deployment in more demanding environments. ARCTRON is also supported by an integrated software suite. Automated control and data acquisition allow discharge parameters, gas composition, magnetic fields, diagnostic timing, and test configuration to be recorded as part of each experiment (STARDAC - Software for Testing, Analysis, Research Data, and Control). The Backend for Experiment Analysis, Storage, and Traceability (BEAST) is a database that provides the infrastructure needed to associate heterogeneous measurements with facility configuration, specimen identity, calibration state, geometry, and analysis provenance. This backend is particularly important for exploratory campaigns, in which many related configurations may be tested, and the value of an individual experiment depends on its connection to earlier and subsequent iterations. Complementary analysis capabilities, including computer-vision-based transient response measurements (arcjetCV), three-dimensional surface reconstruction (STARSCAN), and model-based Bayesian inference (SHIELD), and tomography data analysis (TOMATO, PuMA) can be incorporated when required by a specific hypothesis without becoming the focus of every campaign. The central objective of ARCTRON is therefore not to maximize heat flux or reproduce a complete flight environment. Its purpose is to reduce the cost and time required to ask consequential questions about plasma behavior, plasma-facing materials, diagnostics, and arcjet technology. By providing a controlled environment for rapid reconfiguration, mechanism isolation, quantitative measurement, and iterative engineering, ARCTRON can help mature concepts that would otherwise remain too speculative or too risky for evaluation in production facilities. The resulting knowledge can then guide the design of material models, focus test objectives in larger arcjets, reduce facility-development risk, and improve the physical basis of high-enthalpy ground testing. This work will present the ARCTRON architecture, operating modes, diagnostic suite, and digital experimental workflow. Initial experimental results from the first integrated operation of the facility will be presented, including flow characterization, power limitations, and deployment of the initial diagnostic suite. Ongoing development efforts aimed at catalycity characterization, magnetic plasma control, and advanced optical diagnostics will also be discussed, illustrating how the platform supports rapid iteration from concept to experiment.

experimental diagnostics↗

NASA NTRSDate not supplied

ARCTRON: A Rapid Experimental Proving Ground for TPS Experiments and Arcjet Technology Development

Innovation in high-enthalpy facilities is fundamentally limited by the cost and risk of experimentation. New concepts for plasma control, diagnostics, facility components, and plasma-material interaction often require repeated iterations that are impractical to perform in production arcjets. As a result, promising ideas may remain unexplored or reach operational facilities only after significant development effort. ARCTRON is being developed as a rapid experimental proving ground where new ideas in plasma science, arcjet engineering, diagnostics, and material response can be conceived, tested, and quantitatively evaluated before transition to large-scale facilities. The platform combines radio-frequency (RF) and DC arc plasma generation, externally applied magnetic fields, configurable gas composition, reduced-pressure operation, laser heating, electrical biasing, and modular diagnostic access. These capabilities permit the plasma source, applied forcing, test article, and measurement configuration to be modified independently, allowing individual physical mechanisms to be isolated more readily than in a traditional test environment. One class of investigations addresses fundamental plasma-surface interaction physics. Conventional material tests often expose a specimen simultaneously to convective heating, reactive species, pressure, shear, radiation, and surface-current effects. The resulting material response may be measured accurately, while the contribution of each mechanism remains difficult to identify. ARCTRON is designed to vary these effects selectively. Plasma chemistry can be changed independently through configurable gas mixtures; magnetic fields and electrical biasing can modify charged-particle transport; laser heating can provide a non-plasma thermal input; and pressure, flow, and discharge mode can be varied over a broad operating space. This enables controlled tests of hypotheses involving surface catalycity, reactive-species transport, plasma-assisted oxidation, electromagnetic effects, shear, and the relative contributions of thermal and chemical loading. A second class of investigations enabled by this approach concerns the engineering of high-enthalpy facilities themselves. Arc-heated facilities are limited by electrode erosion, unstable arc attachment, localized heating, and damage to nozzles and other plasma-facing components. ARCTRON provides a lower-cost environment for testing concepts intended to mitigate these limitations. Candidate investigations include the use of applied magnetic fields to alter current paths and reduce plasma interaction with nozzle walls, ExB forcing to introduce controlled plasma rotation, magnetic or geometric approaches for distributing arc attachment, and alternative electrode or discharge configurations intended to reduce erosion and improve stability. Because the platform is reconfigurable, these concepts can be evaluated through repeated design--build--test cycles before they are considered for implementation in operational facilities. The platform also supports the development and validation of diagnostics that may be difficult to introduce initially into a large arcjet. Current and planned measurements include spatially resolved optical emission spectroscopy, electrostatic probes, fast imaging, pyrometry, calorimetry, laser-induced fluorescence, and absorption spectroscopy. These diagnostics are intended not merely to document a nominal operating condition, but to constrain the local plasma state and its relationship to component or material response. The modular facility geometry allows diagnostic concepts to be tested, calibrated, and compared under repeatable conditions before deployment in more demanding environments. ARCTRON is also supported by an integrated software suite. Automated control and data acquisition allow discharge parameters, gas composition, magnetic fields, diagnostic timing, and test configuration to be recorded as part of each experiment (STARDAC - Software for Testing, Analysis, Research Data, and Control). The Backend for Experiment Analysis, Storage, and Traceability (BEAST) is a database that provides the infrastructure needed to associate heterogeneous measurements with facility configuration, specimen identity, calibration state, geometry, and analysis provenance. This backend is particularly important for exploratory campaigns, in which many related configurations may be tested, and the value of an individual experiment depends on its connection to earlier and subsequent iterations. Complementary analysis capabilities, including computer-vision-based transient response measurements (arcjetCV), three-dimensional surface reconstruction (STARSCAN), and model-based Bayesian inference (SHIELD), and tomography data analysis (TOMATO, PuMA) can be incorporated when required by a specific hypothesis without becoming the focus of every campaign. The central objective of ARCTRON is therefore not to maximize heat flux or reproduce a complete flight environment. Its purpose is to reduce the cost and time required to ask consequential questions about plasma behavior, plasma-facing materials, diagnostics, and arcjet technology. By providing a controlled environment for rapid reconfiguration, mechanism isolation, quantitative measurement, and iterative engineering, ARCTRON can help mature concepts that would otherwise remain too speculative or too risky for evaluation in production facilities. The resulting knowledge can then guide the design of material models, focus test objectives in larger arcjets, reduce facility-development risk, and improve the physical basis of high-enthalpy ground testing. This work will present the ARCTRON architecture, operating modes, diagnostic suite, and digital experimental workflow. Initial experimental results from the first integrated operation of the facility will be presented, including flow characterization, power limitations, and deployment of the initial diagnostic suite. Ongoing development efforts aimed at catalycity characterization, magnetic plasma control, and advanced optical diagnostics will also be discussed, illustrating how the platform supports rapid iteration from concept to experiment.

experimental diagnostics↗