Search NASA⌕ Search

SEARCH · Search NASA

Results for “high-throughput inference”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Variable rate neural compression for sparse detector data

Particle colliders produce data at extraordinary rates, posing major challenges for transmission and storage. High-throughput compression algorithms are therefore essential. In the sPHENIX experiment taking data at the Relativistic Heavy Ion Collider, a time projection chamber records three-dimensional (3D) particle trajectories that are highly sparse, making conventional learning-free lossy compression ineffective. Convolutional neural networks have surpassed traditional methods in compression ratio and accuracy. However, they fail to exploit sparsity for efficiency. To address these gaps, we present BCAE-VS, a bicephalous convolutional autoencoder with variable compression ratio for sparse data, which adapts compression to input complexity through key-point identification and sparse convolution. BCAE-VS achieves higher accuracy and compression ratios than prior neural approaches while being orders of magnitude smaller. Moreover, its throughput increases with sparsity—a property not observed in other methods. Although it was developed for collider experiments, BCAE-VS readily extends to other sparse data domains, such as light detection and ranging (LiDAR) sensing and 3D microscopy.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

FPGA-accelerated SpeckleNN with SNL for real-time X-ray single-particle imaging

We present the implementation of a specialized version of our previously published unified embedding model, SpeckleNN, for real-time speckle pattern classification in X-ray Single-Particle Imaging (SPI), using the SLAC Neural Network Library (SNL) on an FPGA platform. This hardware realization transitions SpeckleNN from a prototypic model into a practical edge solution, optimized for running inference near the detector in high-throughput X-ray free-electron laser (XFEL) facilities, such as those found at the Linac Coherent Light Source (LCLS). To address the resource constraints inherent in FPGAs, we developed a more specialized version of SpeckleNN. The original model, which was designed for broader classification across multiple biological samples, comprised ~5.6 million parameters. The new implementation, while reducing the parameter count to 64.6K (a 98.8% reduction), focuses on maintaining the model's essential functionality for real-time operation, achieving an accuracy of 90%. Furthermore, we compressed the latent space from 128 to 50 dimensions. This implementation was demonstrated on the KCU1500 FPGA board, utilizing 71% of available DSPs, 75% of LUTs, and 48% of FFs, with an average power consumption of 9.4W according to the Vivado post-implementation report. The FPGA performed inference on a single image with a latency of 45.015 microseconds at a 200 MHz clock rate. In comparison, running the same inference on an NVIDIA A100 GPU resulted in an average power consumption of ~73W and an image processing latency of around 400 microseconds. Our FPGA-accelerated version of SpeckleNN demonstrated significant improvements, achieving an 8.9 × speedup and a 7.8 × reduction in power consumption compared to the GPU implementation. Key advancements include model specialization and dynamic weight loading through SNL, which eliminates the need for time-consuming FPGA design re-synthesis, allowing fast and continuous deployment of models (re)trained online. These innovations enable real-time adaptive classification and efficient vetoing of speckle patterns, making SpeckleNN more suited for deployment in XFEL facilities. This implementation has the potential to significantly accelerate SPI experiments and enhance adaptability to evolving experimental conditions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Labels as a feature: Network homophily for systematically annotating human GPCR drug-target interactions

Machine learning has revolutionized drug discovery by enabling the exploration of vast, uncharted chemical spaces essential for discovering novel patentable drugs. Despite the critical role of human G protein-coupled receptors in FDA-approved drugs, exhaustive in-distribution drug-target interaction testing across all pairs of human G protein-coupled receptors and known drugs is rare due to significant economic and technical challenges. This often leaves off-target effects unexplored, which poses a considerable risk to drug safety. In contrast to the traditional focus on out-of-distribution exploration (drug discovery), we introduce a neighborhood-to-prediction model termed Chemical Space Neural Networks that leverages network homophily and training-free graph neural networks with labels as features. We show that Chemical Space Neural Networks’ ability to make accurate predictions strongly correlates with network homophily. Thus, labels as features strongly increase a machine learning model’s capacity to enhance in-distribution prediction accuracy, which we show by integrating labeled data during inference. We validate these advancements in a high-throughput yeast biosensing system (3773 drug-target interactions, 539 compounds, 7 human G protein-coupled receptors) to discover novel drug-target interactions for FDA-approved drugs and to expand the general understanding of how to build reliable predictors to guide experimental verification.

Hansson, Frederik G↗
DOE OSTI2025

End-to-end deep learning pipeline for real-time Bragg peak segmentation: from training to large-scale deployment

X-ray crystallography reconstruction, which transforms discrete X-ray diffraction patterns into three-dimensional molecular structures, relies critically on accurate Bragg peak finding for structure determination. As X-ray free electron laser (XFEL) facilities advance toward MHz data rates (1 million images per second), traditional peak finding algorithms that require manual parameter tuning or exhaustive grid searches across multiple experiments become increasingly impractical. While deep learning approaches offer promising solutions, their deployment in high-throughput environments presents significant challenges in automated dataset labeling, model scalability, edge deployment efficiency, and distributed inference capabilities. We present an end-to-end deep learning pipeline with three key components: (1) a data engine that combines traditional algorithms with our peak matching algorithm to generate high-quality training data at scale, (2) a modular architecture that scales from a few million to hundreds of million parameters, enabling us to train large expert-level models offline while deploying smaller, distilled models at the edge, and (3) a decoupled producer-consumer architecture that separates specialized data source layer from model inference, enabling flexible deployment across diverse computing environments. Using this integrated approach, our pipeline achieves accuracy comparable to traditional methods tuned by human experts while eliminating the need for experiment-specific parameter tuning. Although current throughput requires optimization for MHz facilities, our system's scalable architecture and demonstrated model compression capabilities provide a foundation for future high-throughput XFEL deployments.

Wang, Cong↗
DOE OSTI2025

Computer Vision Pipeline for Image Analysis for Freeze‐Fracture Electron Microscopy: Rosette Cellulose Synthase Complexes Case

In materials science, plant biology, agriculture, and environmental research, the automated analysis of high-magnification, complex microscopy images, such as those generated by freeze-fracture electron microscopy (FF-TEM), remains a critical challenge that limits the scalability of data interpretation. We present a deep learning computer vision pipeline for high-throughput detection and morphological characterization analysis of cellulose synthase complexes (CSCs, or rosettes) in FF-TEM images. The pipeline integrates preprocessing, detection, human-in-the-loop verification, and semantic segmentation to quantify features such as rosette diameter and inter-lobe spacing. The approach was trained and tested on a curated dataset of high-resolution FF-TEM micrographs of Physcomitrium patens, expanded via strategic tiling and augmentation to over 650 images. We compare YOLOv8 and YOLOv9 architectures and demonstrate that YOLOv9 achieves superior performance in both localization accuracy (mAP50-95 = 0.854) and inference speed. The resulting distributions revealed biological variability consistent with prior manual studies, validating the approach for high-throughput applications. Our results show that the pipeline achieves human-expert level accuracy while dramatically reducing analysis time, enabling scalable, reproducible structural characterization of intramembrane protein complexes. The pipeline is broadly applicable to other domains requiring precise interpretation of complex microscopy data and establishes a foundation for future artificial intelligence (AI)-assisted workflows in biological imaging.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2000

Genetic Network Inference: From Co-Expression Clustering to Reverse Engineering

Advances in molecular biological, analytical, and computational technologies are enabling us to systematically investigate the complex molecular processes underlying biological systems. In particular, using high-throughput gene expression assays, we are able to measure the output of the gene regulatory network. We aim here to review datamining and modeling approaches for conceptualizing and unraveling the functional relationships implicit in these datasets. Clustering of co-expression profiles allows us to infer shared regulatory inputs and functional pathways. We discuss various aspects of clustering, ranging from distance measures to clustering algorithms and multiple-duster memberships. More advanced analysis aims to infer causal connections between genes directly, i.e., who is regulating whom and how. We discuss several approaches to the problem of reverse engineering of genetic networks, from discrete Boolean networks, to continuous linear and non-linear models. We conclude that the combination of predictive modeling with systematic experimental verification will be required to gain a deeper insight into living organisms, therapeutic targeting, and bioengineering.

Dhaeseleer, Patrik↗
DOE OSTI2026

Finding the missing pieces: filling gaps that impede the translation of omics data into models

High-throughput omics technologies such as DNA sequencing have made the sequencing and computational assembly of microbial genomes recovered from the environment relatively routine. Computational inference of the protein products encoded by these genomes, and the associated biochemical functions, should enable the accurate prediction and modeling of microbial metabolism, organismal interactions, and ecosystem processes. However, a lack of scalable, probabilistic protein annotation tools limits the full potential of modeling for understanding the metabolism and biogeochemical cycles of microbial communities. Our approach to improve inference of protein annotations and metabolic models relied on learning from and emulating expert manual curation, leveraging software engineering and data science best practices to scale up the throughput and accuracy of annotations and metabolic model construction, building software to objectively evaluate different annotation strategies, and more closely linking the protein annotation and metabolic model inference process. Outcomes of this research include several improved or new computational tools, including DRAM (Distilled and Refined Annotation of Metabolism) for annotating microbial genomes with protein function and metabolic traits, CAMPER (Curated Annotations for Microbial Polyphenol Enzymes and Reactions) for annotating key polyphenol metabolisms, EC-Bench for comprehensive and unbiased benchmarking of annotation tools, and several apps available via the DOE Systems Biology Knowledgebase (KBase) for building genome-scale metabolic models. We demonstrate that these tools allow us to scalably annotate and understand thousands of genomes for microbial communities from a variety of systems and test cases, including rivers, thawing permafrost, and gut microbiomes. All of these computational tools are available as open-source software, with most broadly and easily accessible to the scientific community via KBase apps.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Unlocking hidden information in sparse small-angle neutron scattering measurements

Hypothesis Small-Angle Neutron Scattering (SANS) is a powerful technique for studying soft matter systems such as colloids, polymers, and lyotropic phases, providing nanoscale structural insights. However, its effectiveness is limited by low neutron flux, leading to long acquisition times and noisy data. Here, we hypothesize that Bayesian statistical inference using Gaussian Process Regression (GPR) can reconstruct high-fidelity scattering data from sparse measurements by leveraging intensity smoothness and continuity. Experiments and Simulations The method was benchmarked computationally and validated through SANS experiments on various soft matter systems, including wormlike micelles, colloidal suspensions, polymeric structures, and lyotropic phases. GPR-based inference was applied to both experimental and synthetic data to evaluate its effectiveness in noise reduction and intensity reconstruction. Findings GPR significantly enhances SANS data quality and therefore reducing measurement times by up to two orders of magnitude. This cost-effective approach maximizes experimental efficiency, enabling high-throughput studies and real-time monitoring of dynamic systems. It is particularly beneficial for weakly scattering and time-sensitive studies. Beyond SANS, this framework applies to other low-SNR techniques, including laboratory-based small-angle X-ray scattering and various dynamical scattering methods. Furthermore, it offers transformative potential for compact neutron sources, enhancing their viability for structural analysis in resource-limited settings.

Small angle neutron scattering↗
DOE OSTI2026

Patch Hierarchical Attention Transformer for Efficient Particle Jet Tagging

Real-time jet tagging is critical for identifying short-lived particle decays in the high-throughput detectors of the Large Hadron Collider, where real-time trigger systems responsible for deciding which collision events to store impose strict latency and accuracy constraints. While transformer architectures achieve the highest jet tagging accuracy when compute is unconstrained, their quadratic self-attention cost makes inference restrictive on trigger budget. Existing efficient variants reduce the computational cost, but hinder the classification performance. To address this limitation, we introduce the Patch Hierarchical Attention Transformer (PHAT-JeT), which combines two mechanisms: a physics-inspired geometric message-passing module that encodes local detector-plane structure, and a hierarchical patch-based attention scheme that computes exact attention within small particle groups while preserving global context through lightweight patch-token communication. Within a restricted budget, PHAT-JeT achieves state-of-the-art accuracy and background rejection among all resource-constrained jet tagging models on four benchmarks (\textsc{hls4ml}, JetClass, Top Tagging, and Quark--Gluon). Our code is available at https://github.com/aaronw5/PHAT-JeT.

Wang, Aaron [Illinois U., Chicago] (ORCID:00000003↗
DOE OSTI2025

Defect Diffusion Graph Neural Networks for Materials Discovery in High-Temperature Energy Applications

Here, the migration of crystallographic defects dictates material properties and performance for a plethora of technological applications. Density functional theory (DFT)-based nudged elastic band (NEB) calculations are a powerful computational technique for predicting defect migration activation energy barriers, yet they become prohibitively expensive for high-throughput screening of defect diffusivities. Without introducing hand-crafted (i.e., chemistry- or structure-specific) descriptors, we propose a generalized deep learning approach to train surrogate models for NEB energies of vacancy migration by hybridizing graph neural networks with transformer encoders and simply using pristine host structures as input. With sufficient training data, computationally efficient and simultaneous inference of vacancy defect thermodynamics and migration activation energies can be obtained to compute temperature-dependent vacancy diffusivities and to down-select candidates for more thorough DFT analysis or experiments. Thus, as we specifically demonstrate for potential water-splitting materials, candidates with desired defect thermodynamics, kinetics, and host stability properties can be more rapidly targeted from open-source databases of experimentally validated or hypothetical materials.

14 SOLAR ENERGY↗
DOE OSTI2025

Protein-ligand binding affinity prediction using multi-instance learning with docking structures

Recent advances in 3D structure-based deep learning approaches demonstrate improved accuracy in predicting protein-ligand binding affinity in drug discovery. These methods complement physics-based computational modeling such as molecular docking for virtual high-throughput screening. Despite recent advances and improved predictive performance, most methods in this category primarily rely on utilizing co-crystal complex structures and experimentally measured binding affinities as both input and output data for model training. Nevertheless, co-crystal complex structures are not readily available and the inaccurate predicted structures from molecular docking can degrade the accuracy of the machine learning methods. We introduce a novel structure-based inference method utilizing multiple molecular docking poses for each complex entity. Our proposed method employs multi-instance learning with an attention network to predict binding affinity from a collection of docking poses. We validate our method using multiple datasets, including PDBbind and compounds targeting the main protease of SARS-CoV-2. The results demonstrate that our method leveraging docking poses is competitive with other state-of-the-art inference models that depend on co-crystal structures. This method offers binding affinity prediction without requiring co-crystal structures, thereby increasing its applicability to protein targets lacking such data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Conditional diffusion machine-learning framework for mapping valence electron distribution from convergent beam electron diffraction

Quantitative convergent beam electron diffraction (CBED) enables determination of aspherical valence electron distributions through refinement of low-order structure factors, which are highly sensitive to chemical bonding and charge density variations. However, conventional quantitative CBED (QCBED) requires solving a highly nonlinear inverse problem with many coupled parameters, and computationally intensive dynamical diffraction calculations, making it time-consuming and difficult to apply to complex systems. More broadly, reconstructing charge density and orbital electron distribution from diffraction data has long been a central challenge in both x-ray and electron crystallography. Here, in this study, we introduce an artificial-intelligence (AI)-based framework that replaces traditional refinement with a data-driven inverse solver. Using a large synthetic CBED dataset generated by Bloch-wave simulations, we train a conditional diffusion model to directly infer crystal structural parameters and multipole density formalism parameters, and hence valence electron distributions, from CBED patterns alone. By learning from forward simulations across realistic parameter space, the model effectively solves the inverse problem. Compared with direct regression approaches, the diffusion-based framework provides posterior parameter distributions for rigorous uncertainty quantification while preserving quantitative fidelity and reducing analysis time by orders of magnitude. By eliminating the need for external single-crystal x-ray diffraction data and complex nonlinear refinement, this approach enables practical, high-throughput, and in situ quantitative CBED, enabling real-time mapping of valence electron distributions and their correlation with functional responses in quantum and energy materials.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2025

End-to-End Automated Segmentation Framework for Four-Dimensional Scanning Transmission Electron Microscopy Data

Four-dimensional scanning transmission electron microscopy (4D-STEM) is powerful for rapidly characterizing arrays of nanoparticles produced via high-throughput synthesis. However, such 4D-STEM datasets typically contain thousands of nanoparticles, each characterized by thousands of diffraction patterns spatially distributed across the nanoparticle, necessitating efficient and comprehensive analysis. We propose an end-to-end segmentation framework to automatically segment each nanoparticle into regions with distinct composition/orientation of crystal grains, using only the 4D-STEM data. Bragg disk information is extracted in a physics-informed manner from the diffraction patterns at each spatial location and combined with the real space coordinates to form feature vectors. These feature vectors are then used as inputs to a Gaussian mixture model (GMM) to segment the nanoparticle into distinct regions. We also develop two visualization tools based on the GMM outputs to infer the interface transition and the degree of superposition. Our framework comprehensively integrates machine learning tools and physics knowledge, and provides a basis for substantially compressing enormous 4D-STEM datasets, e.g., by replacing the full 4D-STEM dataset for each nanoparticle with only a single set of Bragg disk features for each distinct crystal grain identified in the nanoparticle. In this article, we demonstrate the power of our framework by presenting results for real, complex datasets.

47 OTHER INSTRUMENTATION↗
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 OSTI2026

Segmentation method comparison for residual fiber length measurement across tiled microscopy images

Fiber length distribution (FLD), in part, governs mechanical properties in discontinuous fiber composites, yet manual measurement methods limit the high-throughput characterization needed for materials design optimization. This study compares deep learning segmentation approaches for automated FLD measurement in large-field microscopy, evaluating how method choice affects the microstructural descriptors used in structure-property-processing relationships. A critical challenge is that high-resolution microscopy images (10,000×10,000 pixels) must be tiled for deep learning analysis, fragmenting fibers at boundaries. We demonstrate that segmentation method proves crucial for measurement accuracy. For example, instance segmentation with Slicing Aided Hyper Inference (SAHI) preserves individual fiber integrity across tiles while semantic segmentation prioritizes speed. Comparing against manual measurement of extracted carbon fibers, YOLOv11-SAHI matched manual ground truth (238 μm weighted mean) with 40x speedup (4.5 vs 167 minutes per image). U-Net provides rapid quantification although it is at the cost of reduced accuracy due only reliably measuring stand-alone fibers. Our comparative analysis reveals that instance segmentation with SAHI better preserves length measurements while semantic segmentation prioritizes speed, providing empirical guidance for method selection. The characterization provides essential inputs for mechanical property prediction models and inverse design workflows, accelerating composite materials development cycles.

Additive manufacturing↗
DOE OSTI2025

Bayesian Gaussian process inference for neutron spin echo measurement

Neutron spin echo (NSE) spectroscopy provides unique access to microscopic dynamics, but its application is often constrained by low neutron flux, long acquisition times, and significant noise. Here, we present a Bayesian inference approach based on Gaussian process regression (GPR) to reconstruct high-quality spin echo signals from sparse and noisy data by exploiting correlations in reciprocal space. Benchmarks on synthetic datasets and validation with experimental NSE measurements of dendrimers show that GPR suppresses noise, interpolates missing intensity values, and accommodates irregular observations. The method improves accuracy, shortens acquisition times, and enables high-throughput and real-time studies. Beyond NSE, the framework is broadly applicable to other low signal-to-noise ratio scattering techniques, thereby extending the scope of neutron spectroscopy.

Tung, Chi-Huan [Oak Ridge National Laboratory (ORN↗
DOE OSTI2025

A hybrid neural architecture: Online attosecond x-ray characterization

The emergence of high-repetition-rate x-ray free-electron lasers (XFELs), such as SLAC’s LCLS-II, serves as our canonical example for autonomous controls that necessitate high-throughput diagnostics paired with streaming computational pipelines capable of single-shot analysis with extremely low latency. We present the deterministic characterization with an integrated parallelizable hybrid resolver architecture, a hybrid machine learning framework designed for fast, accurate analysis of XFEL diagnostics using angular streaking-based sinogram images. This architecture integrates convolutional neural networks and bidirectional long short-term memory models to denoise input, identify x-ray sub-spike features, and extract sub-spike relative delays with sub-30 attosecond temporal resolution. Deployed on low-latency hardware, it achieves over 10 kHz throughput with 168.3 μs inference latency, indicating scalability to 14 kHz with field-programmable gate array integration. By transforming regression tasks into classification problems and leveraging optimized error encoding, we achieve high precision with low-latency performance that is critical for real-time streaming event selection and experimental control feedback signals. This represents a key development in real-time control pipelines for next-generation autonomous science, generally, and high repetition-rate x-ray experiments in particular.

Accelerator Physics (physics.acc-ph)↗
DOE OSTI2025

Integrating and Characterizing HPC Task Runtime Systems for hybrid AI-HPC workloads

Scientific workflows increasingly involve both HPC and machine-learning tasks, combining MPI-based simulations, training, and inference in a single execution. Launchers such as Slurm’s srun constrain concurrency and throughput, making them unsuitable for dynamic and heterogeneous workloads. We present a performance study of RADICAL-Pilot (RP) integrated with Flux and Dragon, two complementary runtime systems that enable hierarchical resource management and high-throughput function execution. Using synthetic and production-scale workloads on Frontier, we characterize the task execution properties of RP across runtime configurations. RP+Flux sustains up to 930 tasks/s, and RP+Flux+Dragon exceeds 1,500 tasks/s with over 99.6% utilization. In contrast, srun peaks at 152 tasks/s and degrades with scale, with utilization below 50%. For IMPECCABLE.v2 drug discovery campaign, RP+Flux reduces makespan by 30–60% relative to srun/Slurm and increases throughput more than four times on up to 1,024. These results demonstrate hybrid runtime integration in RP as a scalable approach for hybrid AI-HPC workloads.

HPC-AI↗