Search NASA⌕ Search

SEARCH · Search NASA

Results for “Cloud Architectures”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Confidentiality-preserving machine learning algorithms for soft-failure detection in optical communication networks

Automated fault management is at the forefront of next-generation optical communication networks. The increase in complexity of modern networks has triggered the need for programmable and software-driven architectures to support the operation of agile and self-managed systems. In these scenarios, the European Telecommunications Standards Institute zero-touch network and service management approach is imperative. The need for machine learning algorithms to process the large volume of telemetry data brings safety concerns as distributed cloud-computing solutions become the preferred approach for deploying reliable communication network automation. This paper’s contribution is twofold. First, we propose a simple yet effective method to guarantee the confidentiality of the telemetry data based on feature scrambling. The method allows the operation of third-party computational services without direct access to the full content of the collected data. Additionally, the effectiveness of four unsupervised machine learning algorithms for soft-failure detection is evaluated when applied to the scrambled telemetry data. The methods are based on factor analysis, principal component analysis, nonlinear principal component analysis, and singular value decomposition. Most dimensionality reduction algorithms have the common property that they can maintain similar levels of fault classification performance while hiding the data structure from unauthorized access. Evaluations of the proposed algorithms demonstrate this capability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Data-driven emulation of modal aerosol microphysics via neural operator-based modeling

The complexity and the small characteristic scales of aerosol microphysical processes pose a big challenge for accurate and efficient Earth system simulations at regional and global scales. In this work, we construct and evaluate a surrogate model: the aerosol deep operator network (ADON), a physics-inspired dual-net architecture for emulating the aerosol microphysics parameterization suite in the version 2 of the Energy Earth System Model (E3SMv2). The current version of the surrogate model is trained on a dataset comprising 9.8 million samples obtained from a global E3SMv2 simulation with the horizontal resolution of about one degree under cloud-free conditions. Incorporating domain spatial and temporal coordinates, as well as principle components extracted from training data, the dual-net surrogate model effectively captures the intricate representations of aerosol and the relationship with atmospheric state variables, achieving an R-squared score over $$95.7\%$$ for all the lognormal aerosol modes in the extrapolated regime. The validated model provides feature importance of input variables and their impact on the predictive capacity of the surrogate model in relation to the E3SM. The computational cost of online inference time deployed on CPUs and GPUs with lower precisions highlights ADON’s efficiency and potential in robust predictive modeling for large-scale Earth system computations.

Bai, Zhe↗
DOE OSTI2024

A Roadmap for a Lightning Modeling Grand Challenge

This document is a roadmap for building an interconnected model of the physical processes that produce a lightning discharge, and its observable optical and radio signals. We call this a Lightning Modeling Grand Challenge, recognizing that significant effort and coordination of human and financial resources is required to realize the capability. The roadmap serves to outline the coordination of resources necessary to enable stitching together existing knowledge and model components to make a lightning prediction, and to test these predictions with observations. Such a capability does not currently exist. The roadmap is motivated not only by a spirit of scientific inquiry, but by practical challenges faced by US Federal and societal stakeholders. Advancements in lightning observations have outpaced our tests of integrated understanding, leaving many stakeholders unsure how to design their missions to properly detect and discriminate lightning, and unsure how to apply the sometimes-disagreeing lightning signals from diverse instruments. The time is right to connect existing theories and models to support stakeholders in understanding the signals they observe, for needs as diverse as climate monitoring, national security, weather forecasting, public safety, and protection of natural and built environments. The roadmap’s two main technical sections describe the components of a linked physical model, followed by a description of models of lightning signals and sensors that are driven by outputs from the physical model. The goal is to predict the time-varying physical properties of lightning that are self-consistent with the thunderstorm’s structure and dynamics. These lightning signals then propagate through the storm, with realistic dispersion and attenuation, to receivers on the ground or in space. At a high level, the model begins with weather (cloud) model output, including explicit prediction of the electrification of cloud particles. The cloud’s electrical structure drives a model of lightning physics, from initiation, through channel development, and discharges along those channels. Key lightning parameters, such as the temperature and currents in the channel, and their space and time distribution, are then used to produce optical and electromagnetic signal sources that propagate to modeled receivers. This architecture therefore generates a dataset suitable for comparison to existing and envisioned observing systems. The need for additional measurements and field campaigns to support model development is described. In each model sub-component, inputs, outputs, uncertainties, evaluation methods, and next steps are summarized, interleaved with references to the scientific literature. Identifying boundaries between the model sub-components aids in segmenting an integrated, complex model into practical work packages and system sub-components, allowing a diverse team to contribute and maintain the system. We estimate that at least five years of effort and a $\$$10M initial investment is necessary to make a significant step forward. Mechanisms to facilitate community coordination, including annual workshops and open-source code repositories, are described.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

DOE Repository Metadata Profile (DRMP): A Metadata Framework for Advancing Interoperability and AI Readiness Across Scientific Repositories

The Department of Energy (DOE) funds a diverse and distributed ecosystem of repositories that steward scientific data, publications, and software across its research programs, user facilities, and national laboratories. While significant progress has been made in standardizing dataset-level metadata, the metadata describing repositories themselves (their identity, governance, access interfaces, policies, and technical capabilities) remains inconsistent and fragmented across DOE-funded systems. This variability limits discoverability, interoperability, automated validation, and AI-driven analysis, all of which are increasingly essential for modern scientific workflows. To address this gap, the DOE Data Curation Working Group (DCWG) developed the DOE Repository Metadata Profile (DRMP). The DRMP is a practical, community-driven framework that defines how repositories can describe themselves in a consistent, machine-actionable, and scalable manner. The DRMP is not a new metadata schema. Instead, it is a mapping profile and structured element set capturing the essential characteristics of DOE repositories. It harmonizes repository-level metadata across six widely adopted community schemas: RE3Data; DCAT-US v3; Schema.org; Dublin Core; DataCite 4.6; and PREMIS 3.0. This harmonization eliminates reinvention and enables interoperability within DOE and across the broader scientific ecosystem. A core objective of the DRMP is to reduce burden on repositories by allowing them to reuse their existing metadata through a Rosetta-style crosswalk rather than redesigning local implementations. The profile introduces a three-level conformance model that supports incremental adoption: • Level 1 – Minimum Viable Record (MVR): foundational identification elements required for workflows, project registration, and basic repository presence. • Level 2 – Interoperable: structured metadata enabling alignment with national and international discovery systems. • Level 3 – AI-Ready: enhanced provenance, policy transparency, fixity, semantic context, and capabilities that support automated reasoning, model training governance, and machine-assisted curation. To support implementation, the DRMP includes JSON Schema definitions, OpenAPI patterns, and MCP templates that allow repositories to publish machine-readable metadata directly within existing platforms. These resources are modular and lightweight, enabling adoption without major architectural change. Adopting the DRMP enables repositories to: • Enhance discoverability and interoperability by aligning identifiers, classifications, and descriptive elements across widely used schema standards. • Support federated discovery and cross-registration across DOE systems, Data.gov, and international catalogs. • Enable AI agents and workflow orchestration systems to interpret repository-level metadata within the American Science Cloud (AmSC) through Model Context Protocol (MCP)-based context publication. • Demonstrate alignment with DOE’s open science, stewardship, and FAIR data priorities. This guidance represents a community-driven step forward. Through voluntary adoption and continued feedback, the DRMP advances a cohesive, machine-actionable description of DOE repositories that supports FAIR data practices, preparing the infrastructure for AI-enabled research, and strengthening the discoverability and reuse of DOE’s scientific outputs.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2022

The Edge of Exploration: An Edge Storage and Computing Framework for Ambient Noise Seismic Interferometry Using Internet of Things Based Sensor Networks

Recent technological advances have reduced the complexity and cost of developing sensor networks for remote environmental monitoring. However, the challenges of acquiring, transmitting, storing, and processing remote environmental data remain significant. The transmission of large volumes of sensor data to a centralized location (i.e., the cloud) burdens network resources, introduces latency and jitter, and can ultimately impact user experience. Edge computing has emerged as a paradigm in which substantial storage and computing resources are located at the “edge” of the network. In this paper, we present an edge storage and computing framework leveraging commercially available components organized in a tiered architecture and arranged in a hub-and-spoke topology. The framework includes a popular distributed database to support the acquisition, transmission, storage, and processing of Internet-of-Things-based sensor network data in a field setting. We present details regarding the architecture, distributed database, embedded systems, and topology used to implement an edge-based solution. Lastly, a real-world case study (i.e., seismic) is presented that leverages the edge storage and computing framework to acquire, transmit, store, and process millions of samples of data per hour.

58 GEOSCIENCES↗
DOE OSTI2025

Introduction: Neuromorphic Materials

The explosive growth in data collection and the need to process it efficiently, as well as the desire to automate increasingly complex tasks in transportation, medical care, manufacturing, security and many other fields have motivated a growing interest in neuromorphic computing. Unlike the binary, transistorbased ON/OFF logic gates and separate logic and memory functionalities employed in digital computing, neuromorphic computing is inspired by animal brains that use interconnected synapses and neurons to perform processing, storage and transmission of information at the same location, while only consuming ~20 W or less of power. Motivated by the brain’s efficiency, adaptability, self-learning and resiliency qualities, neuromorphic computing can be broadly defined as an approach to processing and storing information using hardware and algorithms inspired by models of biological neural systems. Present research in neuromorphic computing encompasses approaches that vary significantly in their degree of neuro-inspiration, from systems that only incorporate features such as asynchronous, event-driven operation or use crossbar arrays of non-volatile memory (NVM) elements to accelerate deep neural networks (DNNs), to designs that embrace the extreme parallelism, sparsity, reconfigurability, adaptability, complexity and stochasticity observed in nervous systems. The term ‘neuromorphic’ computing is often credited to Carver Mead, who in the 1980s investigated Si-based analog electronics to replicate functions of the animal retina. Earlier important advances in this field include the work of Frank Rosenblatt, who proposed the concept of the perceptron, Bernard Widrow, who used this concept to build one of the first analog neural networks, the Adaline and many other researchers (see ref. 6 for an historical perspective on neuromorphic computing). With the recent increase in the use of artificial intelligence and large language models, and rising concerns over the associated energy costs, interest in neuromorphic hardware has expanded rapidly. According to some estimates, driven largely by the drastic growth in the training use of artificial intelligence (AI) models using the current computing architectures, the energy cost of computing is projected to reach the energy supply worldwide by 2045. Furthermore, while this is not a realistic outcome, it means that, if more efficient computing technologies are not developed -- soon -- the world will soon become one where demand for energy and market constraints limit the continued increase of societal access to AI and cloud services from data centers. Data centers used for training and use of these models consume hundreds of terawatt hours of electricity, already past 4% of the US electricity demand.

Circuits↗
DOE OSTI2021

MemXCT: Design, Optimization, Scaling, and Reproducibility of X-Ray Tomography Imaging

Here, this work extends our previous research entitled "MemXCT: Memory-centric X-ray CT Reconstruction with Massive Parallelization" that was originally published at SC19 conference (Hidayetoglu et al., 2019) with reproducibility of the computational imaging performance. X-ray computed tomography (XCT) is regularly used at synchrotron light sources to study the internal morphology of materials at high resolution. However, experimental constraints, such as radiation sensitivity, can result in noisy or undersampled measurements. Further, depending on the resolution, sample size and data acquisition rates, the resulting noisy dataset can be in the order of terabytes. Advanced iterative reconstruction techniques can produce high-quality images from noisy measurements, but their computational requirements have made their use an exception rather than the rule. We propose a novel memory-centric approach that avoids redundant computations at the expense of additional memory complexity. We develop a memory-centric iterative reconstruction system, MemXCT, that uses an optimized SpMV implementation with two-level pseudo-Hilbert ordering and multi-stage input buffering. We evaluate MemXCT on various supercomputer architectures involving KNL and GPU. MemXCT can reconstruct a large (11Kx11K) mouse brain tomogram in 10 seconds using 4096 KNL nodes (256K cores). The results presented in our original article at the SC19 were based on large-scale supercomputing resources. The MemXCT application was selected for the Student Cluster Competition (SCC) Reproducibility Challenge and evaluated on a variety of cloud computing resources by universities around the world in the SC20 conference. We summarize the results of the top-ranked SCC Reproducibility Challenge teams and identify the most pertinent measures for ensuring the reproducibility of our experiments in this article.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

If We Build Them, They Will Run: Automated HPC Apps Deployment and Profiling with eBPF in Cloud

The high performance computing (HPC) community is in a period of transition. The rise of AI/ML coupled with a changing landscape of resources deems portability a new metric of performance, and methods to move between on-premises and cloud environments and assess compatibility are paramount. Here we design and test a strategy for bridging the gap between traditional HPC and Kubernetes environments – first containerizing applications, providing automated orchestration to run studies, and packaging the setup with automated means to assess performance using low overhead eXtended Berkeley Packet Filter (eBPF) programs. We first assess different designs for eBPF collection, demonstrating a tradeoff between number of programs deployed on a node and overhead added. We develop 5 low overhead eBPF programs that combine with streaming ML models to assess CPU, futex, TCP, shared memory, and file access across four different builds of an HPC application for CPU and GPU. We use eBPF data to generate insights into the possible underlying etiology of scaling issues. We then assess compatibility of a well-known benchmark, HPCG, across matrices of micro-architectures and optimization levels (217 containers across 24 instance types and over 7500 runs). We provide to the community 30 applications to deploy in our automated setup and perform a scaling study from 4 to a maximum of 256 nodes for both CPU and GPU applications. Finally, we use our gained knowledge about performance to generate compatibility artifacts that are used by a newly developed Kubernetes controller to intelligently select instance type based on optimizing a figure of merit. Along with insights to scaling in this environment with a collection of applications and templates to work from, we provide an overall strategy for approaching HPC application deployment and image selection based on compatibility in cloud.

Computer science↗
DOE OSTI2026

Deep Learning and Photogrammetric Reconstruction for Automated Crack Detection and Dimensional Measurement in Mining Operations

Surface crack detection and dimensional measurement at active mining sites present significant safety and operational challenges. Manual inspection methods are labor-intensive, spatially incomplete, and expose personnel to hazardous environments, while existing automated approaches have been developed primarily for concrete civil infrastructure and have not been validated on the complex, variable surfaces characteristic of mining environments. This dissertation presents an automated pipeline that integrates deep learning semantic segmentation with Structure-from-Motion photogrammetry to detect surface cracks and measure their aperture, length, and vertical displacement from standard RGB imagery acquired during routine Uncrewed Aerial Vehicle (UAV) survey operations, without requiring additional sensor hardware or manual measurement. The pipeline combines a U-Net architecture with an EfficientNet-B0 encoder, pretrained on the SDNET2018 concrete crack dataset and fine-tuned on a mining-specific dataset spanning laboratory concrete specimens, coal refuse impoundment embankments, and post-blast limestone quarry benches. Photogrammetric reconstruction is performed using COLMAP Structure-from-Motion and Multi-View Stereo, with crack segmentation masks projected into the reconstructed point cloud to enable three-dimensional vertical displacement measurement through local plane fitting and bimodal surface detection. The pipeline was validated across 36 controlled laboratory specimens at three imaging distances and four vertical displacement levels, achieving aperture measurement RMSE of 0.047 cm and R² of 0.954, and vertical displacement RMSE of 0.140 cm and R² of 0.966, against independent caliper measurements. Field application at a coal refuse impoundment in southwestern Pennsylvania detected 71 crack components across the embankment crest, with a dominant longitudinal crack exhibiting aperture values reaching 28 cm and a 95th percentile vertical displacement of 35.53 cm, consistent in magnitude and spatial distribution with simultaneously acquired LiDAR-derived estimates. Application across four post-blast limestone quarry bench datasets in California successfully characterized blast-induced fracture networks at ground sampling distances ranging from 0.59 to 1.23 cm/pixel, with detected crack geometries physically consistent with observable surface conditions at each site. The results demonstrate that deep learning-based crack detection and photogrammetric measurement can be integrated into routine UAV inspection workflows at mining sites, providing repeatable, scalable, and quantitative crack characterization across surface types, crack scales, and displacement magnitudes not previously addressed in the literature. The pipeline requires no dedicated surveying equipment beyond the UAV platforms already deployed at mine sites for survey and monitoring purposes, supporting practical adoption within existing operational workflows.

Crack detection, Dimensional Measurement↗
DOE OSTI2022

Short-term solar radiation forecast using total sky imager via transfer learning

Ground-based sky cameras, which capture hemispherical images, have been extensively used for localized monitoring of clouds. This paper proposes a short-term forecasting approach based on transfer learning using Total Sky-Imager (TSI) images of the Southern Great Plains (SGP) site obtained from the Atmospheric Radiation Measurement (ARM) dataset. An accurate estimation of solar irradiance using TSI is key for short-term solar energy generation forecasting and optimal energy consumption planning. We make use of deep neural network architectures such as AlexNet and ResNet-101 to extract the underlying deep convolution features from TSI images and then train using an ensemble learning approach to model and forecast solar radiation. We demonstrate the performance of the proposed approach by showcasing the best and worst cases. Thus, the transfer learning approach significantly reduces the time and resources required for modeling solar radiation. We outperform with reference to another state-of-art technique for solar modeling using TSI images at different forecast lead times.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Demonstrating the data center as a flexible grid asset using a C-HIL setup

Increasing data center demand is outpacing grid infrastructure development. Artificial intelligence workloads and hyperscale cloud growth are creating unprecedented demand for power, while traditional grid expansion faces multiyear development timelines. Verrus is developing an innovative datacenter solution for this challenge, data centers that act as active grid-supportive assets rather than passive loads. Our approach integrates a novel grid-aware power flow management system with battery energy storage systems(BESS) into a microgrid-controlled, medium-voltage power distribution architecture that delivers critical capabilities, such as: * Fast response to grid disturbances such over/ under voltage or over/ under frequency * Demand flexibility that can service requests from the utility within 10 s * Uninterrupted transition to islanded operation during grid outages * Continuous uptime assurance for compute loads while maintaining all customer service level agreements. Through Verrus' strategic partnership with the National Renewable Energy Laboratory (NREL), these capabilities were validated using NREL's Advanced Research on Integrated Energy Systems (ARIES) virtual emulation environment to model a 70-MW grid-interactive data center. This paper outlines the design, methodology, and results of this emulated deployment, demonstrating that data centers can provide both critical load resilience and ancillary grid support without compromising uptime requirements. Specifically, we present a digital real time simulation of a 70 MW data center integrated with a physical microgrid controller, and demonstrate the data center response in the event of a grid voltage and frequency event, utility demand response request and utility outage.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Towards accelerating particle-resolved direct numerical simulation with neural operators

In this paper, we present our ongoing work aimed at accelerating a particle-resolved direct numerical simulation model designed to study aerosol–cloud–turbulence interactions. The dynamical model consists of two main components—a set of fluid dynamics equations for air velocity, temperature, and humidity, coupled with a set of equations for particle (i.e., cloud droplet) tracing. Rather than attempting to replace the original numerical solution method in its entirety with a machine learning (ML) method, we consider developing a hybrid approach. We exploit the potential of neural operator learning to yield fast and accurate surrogate models and, in this study, develop such surrogates for the velocity and vorticity fields. We discuss results from numerical experiments designed to assess the performance of ML architectures under consideration as well as their suitability for capturing the behavior of relevant dynamical systems.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Vulcan Test Platform: Demonstrating the Data Center as a Flexible Grid Asset

Explosive data center demand is outpacing grid infrastructure development. AI workloads and hyperscale cloud growth are creating unprecedented power requirements, while traditional grid expansion faces multi-year development timelines, regulatory hurdles, and decarbonization challenges. Sidewalk Infrastructure Partners recognized this impending crisis years ago and founded Verrus to develop an innovative solution: data centers that function as grid assets rather than passive loads. The Verrus approach integrates proprietary grid-aware controls with battery energy storage systems (BESS) in a medium-voltage architecture that delivers three critical capabilities: Fast-responding demand flexibility that can service requests from the utility within 10 seconds, Uninterrupted transition to islanded operation during grid disturbances, Continuous uptime assurance while maintaining all customer service level commitments Through Verrus' strategic partnership with the National Renewable Energy Laboratory (NREL), we validated these capabilities on Vulcan, a 70 MW utility-scale test platform powered by NREL's ARIES Virtual Emulation Environment. This deployment-ready technology has successfully demonstrated that Verrus data centers can deliver meaningful grid services while maintaining mission-critical reliability. This technical report outlines the design, methodology, and results of this emulated deployment, demonstrating that data centers can play a pivotal role in enhancing grid flexibility and reliability, without sacrificing service level guarantees.

24 POWER TRANSMISSION AND DISTRIBUTION↗
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 OSTI2020

A Dynamic Reduction Network for Point Clouds

Classifying whole images is a classic problem in machine learning, and graph neural networks are a powerful methodology to learn highly irregular geometries. It is often the case that certain parts of a point cloud are more important than others when determining overall classification. On graph structures this started by pooling information at the end of convolutional filters, and has evolved to a variety of staged pooling techniques on static graphs. In this paper, a dynamic graph formulation of pooling is introduced that removes the need for predetermined graph structure. It achieves this by dynamically learning the most important relationships between data via an intermediate clustering. The network architecture yields interesting results considering representation size and efficiency. It also adapts easily to a large number of tasks from image classification to energy regression in high energy particle physics.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Portable Acceleration of CMS Computing Workflows with Coprocessors as a Service

Computing demands for large scientific experiments, such as the CMS experiment at the CERN LHC, will increase dramatically in the next decades. To complement the future performance increases of software running on central processing units (CPUs), explorations of coprocessor usage in data processing hold great potential and interest. Coprocessors are a class of computer processors that supplement CPUs, often improving the execution of certain functions due to architectural design choices. We explore the approach of Services for Optimized Network Inference on Coprocessors (SONIC) and study the deployment of this as-a-service approach in large-scale data processing. In the studies, we take a data processing workflow of the CMS experiment and run the main workflow on CPUs, while offloading several machine learning (ML) inference tasks onto either remote or local coprocessors, specifically graphics processing units (GPUs). With experiments performed at Google Cloud, the Purdue Tier-2 computing center, and combinations of the two, we demonstrate the acceleration of these ML algorithms individually on coprocessors and the corresponding throughput improvement for the entire workflow. This approach can be easily generalized to different types of coprocessors and deployed on local CPUs without decreasing the throughput performance. We emphasize that the SONIC approach enables high coprocessor usage and enables the portability to run workflows on different types of coprocessors.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Spatially Aware Linear Transformer (SAL-T) for Particle Jet Tagging

Transformers are very effective in capturing both global and local correlations within high-energy particle collisions, but they present deployment challenges in high-data-throughput environments, such as the CERN LHC. The quadratic complexity of transformer models demands substantial resources and increases latency during inference. In order to address these issues, we introduce the Spatially Aware Linear Transformer (SAL-T), a physics-inspired enhancement of the linformer architecture that maintains linear attention. Our method incorporates spatially aware partitioning of particles based on kinematic features, thereby computing attention between regions of physical significance. Additionally, we employ convolutional layers to capture local correlations, informed by insights from jet physics. In addition to outperforming the standard linformer in jet classification tasks, SAL-T also achieves classification results comparable to full-attention transformers, while using considerably fewer resources with lower latency during inference. Experiments on a generic point cloud classification dataset (ModelNet10) further confirm this trend. Our code is available at https://github.com/aaronw5/SAL-T4HEP.

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