Search NASA⌕ Search

SEARCH · Search NASA

Results for “evaluation datasets”

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 379 records · Page 21

Optimal Client Sampling in Federated Learning with Client-level Heterogeneous Differential Privacy

Federated Learning with client-level differential privacy (DP) provides a promising framework for collaboratively training models while rigorously protecting clients’ privacy. However, classic approaches like DP-FedAvg struggle when clients have heterogeneous privacy requirements, as they must uniformly enforce the strictest privacy level across all clients, leading to excessive DP noise and significant degradation in model utility. Existing methods to improve the model utility in such heterogeneous privacy settings often assume a trusted server and are largely heuristic, resulting in suboptimal performance and lacking strong theoretical foundations. Here, in this work, we address these challenges under a practical attack model where both clients and the server are honest-but-curious. We propose GDPFed, which partitions clients into groups based on their privacy budgets and achieves client-level DP within each group to reduce the privacy budget waste and hence improve the model utility. Based on the privacy and convergence analysis of GDPFed, we find that the magnitude of DP noise depends on both model dimensionality and the per-group client sampling ratios. To further improve the performance of GDPFed, we introduce GDPFed+, which integrates model sparsification to eliminate unnecessary noise and optimizes per-group client sampling ratios to minimize convergence error. Extensive empirical evaluations on multiple benchmark datasets demonstrate the effectiveness of GDPFed+, showing substantial performance gains compared with state-of-the-art methods.

Xu, Jiahao [Univ. of Nevada, Reno, NV (United Stat↗

Open Source Synergy: Developing and Validating PMU Data Analysis Techniques Using Open Source Tools and Datasets

This paper presents an exploration into the development and validation of data analysis approaches for Phasor Measurement Units (PMUs) using open-source datasets and tools. Various methods for event detection, event classification, frequency response, and oscillation analysis were tested. We leverage the capabilities of Archive Walker (AW), the Frequency Response Analysis Tool (FRAT), and the Oscillation Baselining and Analysis Tool (OBAT), all open-source tools, for efficient processing and analysis of synchrophasor data. The open-source Transmission Signature Library (TSL) dataset was employed as a dataset for a comprehensive evaluation to assess the performance and reliability of the proposed methods.

PMU, event analysis, oscillation, Frequency Respon↗

An Accurate, Error-Tolerant, and Energy-Efficient Neural Network Inference Engine Based on SONOS Analog Memory

In this work, we demonstrate SONOS (silicon-oxide-nitrideoxide- silicon) analog memory arrays that are optimized for neural network inference. The devices are fabricated in a 40nm process and operated in the subthreshold regime for in-memory matrix multiplication. Subthreshold operation enables low conductances to be implemented with low error, which matches the typical weight distribution of neural networks, which is heavily skewed toward near-zero values. This leads to high accuracy in the presence of programming errors and process variations. We simulate the end-to-end neural network inference accuracy, accounting for the measured programming error, read noise, and retention loss in a fabricated SONOS array. Evaluated on the ImageNet dataset using ResNet50, the accuracy using a SONOS system is within 2.16% of floating-point accuracy without any retraining. The unique error properties and high On/Off ratio of the SONOS device allow scaling to large arrays without bit slicing, and enable an inference architecture that achieves 20 TOPS/W on ResNet50, a >10× gain in energy efficiency over state-of-the-art digital and analog inference accelerators.

97 MATHEMATICS AND COMPUTING↗

Patient-bounded extrapolation using low-dose priors for volume-of-interest imaging in C-arm CT

Purpose: Three-dimensional (3D) volume-of-interest (VOI) imaging with C-arm systems provides anatomical information in a predefined 3D target region at a considerably low x-ray dose. However, VOI imaging involves laterally truncated projections from which conventional reconstruction algorithms generally yield images with severe truncation artifacts. Heuristic based extrapolation methods, e.g., water cylinder extrapolation, typically rely on techniques that complete the truncated data by means of a continuity assumption and thus appear to be ad-hoc. It is our goal to improve the image quality of VOI imaging by exploiting existing patient-specific prior information in the workflow. Methods: A necessary initial step prior to a 3D acquisition is to isocenter the patient with respect to the target to be scanned. To this end, low-dose fluoroscopic x-ray acquisitions are usually applied from anterior–posterior (AP) and medio-lateral (ML) views. Based on this, the patient is isocentered by repositioning the table. In this work, we present a patient-bounded extrapolation method that makes use of these noncollimated fluoroscopic images to improve image quality in 3D VOI reconstruction. The algorithm first extracts the 2D patient contours from the noncollimated AP and ML fluoroscopic images. These 2D contours are then combined to estimate a volumetric model of the patient. Forward-projecting the shape of the model at the eventually acquired C-arm rotation views gives the patient boundary information in the projection domain. In this manner, we are in the position to substantially improve image quality by enforcing the extrapolated line profiles to end at the known patient boundaries, derived from the 3D shape model estimate. Results: The proposed method was evaluated on eight clinical datasets with different degrees of truncation. The proposed algorithm achieved a relative root mean square error (rRMSE) of about 1.0% with respect to the reference reconstruction on nontruncated data, even in the presence of severe truncation, compared to a rRMSE of 8.0% when applying a state-of-the-art heuristic extrapolation technique. Conclusions: The method we proposed in this paper leads to a major improvement in image quality for 3D C-arm based VOI imaging. It involves no additional radiation when using fluoroscopic images that are acquired during the patient isocentering process. The model estimation can be readily integrated into the existing interventional workflow without additional hardware.

60 APPLIED LIFE SCIENCES↗

SU-F-P-49: Comparison of Mapcheck 2 Commission for Photon and Electron Beams

Purpose: We will investigate the performance variation of the MapCheck2 detector array with different array calibration and dose calibration pairs from different radiation therapy machine. Methods: A MapCheck2 detector array was calibrated on 3 Elekta accelerators with different energy of photon (6 MV, 10 MV, 15 MV and 18 MV) and electron (6 MeV, 9 MeV, 12 MeV, 15 MeV, 18 MeV and 20 MeV) beams. Dose calibration was conducted by referring a water phantom measurement following TG-51 protocol and commission data for each accelerator. A 10 cm × 10 cm beam was measured. This measured map was morphed by applying different calibration pairs. Then the difference was quantified by comparing the doses and similarity using gamma analysis of criteria (0.5 %, 0 mm). Profile variation was evaluated on a same dataset with different calibration pairs. The passing rate of an IMRT QA planar dose was calculated by using 3 mm and 3% criteria and compared with respect to each calibration pairs. Results: In this study, a dose variation up to 0.67% for matched photons and 1.0% for electron beams is observed. Differences of flatness and symmetry can be as high as 1% and 0.7% respectively. Gamma analysis shows a passing rate ranging from 34% to 85% for the standard 10 × 10 cm field. Conclusion: Our work demonstrated that a customized array calibration and dose calibration for each machine is preferred to fulfill a high standard patient QA task.

60 APPLIED LIFE SCIENCES↗

Machine Learning to Improve Retrieval by Category in Big Volunteered Geodata

Nowadays, Volunteered Geographic Information (VGI) is commonly used in research and practical applications. However, the quality assurance of such a geographic data remains a problem. In this study we use machine learning and natural language processing to improve record retrieval by category (e.g. restaurant, museum, etc.) from Wikimapia Points of Interest data. We use textual information contained in VGI records to evaluate its ability to determine the category label. The performance of the trained classifier is evaluated on the complete dataset and then is compared with its performance on regional subsets. Preliminary analysis shows significant difference in the classifier performance across the regions. Such geographic differences will have a significant effect on data enrichment efforts such as labeling entities with missing categories.

Sorokine, Alexandre↗

CausalFS

SAND2021-4723 O CausalFS is a Python package that evaluates causal relationships between dataset features and the metric to be predicted. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Nguyen, Bernard↗

HydraGNN_GFM_FineTuning4Materials v1.0

This repository enables fine-tuning of the HydraGNN Predictive GFM 2026 — an open-source ensemble of pre-trained graph foundation models for atomistic materials modeling, developed at Oak Ridge National Laboratory. The GFM 2026 is freely available and downloadable via Globus from the OLCF Data Constellation (DOI: 10.13139/OLCF/2562660). Starting from these pre-trained weights, this repository provides a complete transfer learning pipeline for adapting the GFM ensemble to domain-specific molecular and materials property prediction tasks. It includes: 1) Utilities for ensemble fine-tuning with task-specific output heads 2) Example pipelines for eight widely-used materials and molecular datasets 3) Tools for model adaptation and head configuration 4) Data preprocessing utilities for each supported dataset 5) Benchmarking and evaluation scripts

Ungerboeck, Linda↗

Deep machine learning-assisted multiphoton microscopy to reduce light exposure and expedite imaging

Two-photon excitation fluorescence (2PEF) allows imaging of tissue up to about one millimeter in thickness. Typically, reducing fluorescence excitation exposure reduces the quality of the image. However, using deep learning super resolution techniques, these low-resolution images can be converted to high-resolution images. This work explores improving human tissue imaging by applying deep learning to maximize image quality while reducing fluorescence excitation exposure. We analyze two methods: a method based on U-Net, and a patch-based regression method. Both methods are evaluated on a skin dataset and an eye dataset. The eye dataset includes 1200 paired high power and low power images of retinal organoids. The skin dataset contains multiple frames of each sample of human skin. High-resolution images were formed by averaging 70 frames for each sample and low-resolution images were formed by averaging the first 7 and 15 frames for each sample. The skin dataset includes 550 images for each of the resolution levels. We track two measures of performance for the two methods: mean squared error (MSE) and structural similarity index measure (SSIM). For the eye dataset, the patches method achieves an average MSE of 27,611 compared to 146,855 for the U-Net method, and an average SSIM of 0.636 compared to 0.607 for the U-Net method. For the skin dataset, the patches method achieves an average MSE of 3.768 compared to 4.032 for the U-Net method, and an average SSIM of 0.824 compared to 0.783 for the U-Net method. Despite better performance on image quality, the patches method is worse than the U-Net method when comparing the speed of prediction, taking 303 seconds to predict one image compared to less than one second for the U-Net method.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

VECTOR Phase 1 Dataset: CAV Trajectory and Energy Consumption Records

This dataset contains benchmark experimental data from Phase 1 of the VECTOR project, focusing on the energy impact of CAV hardware components. The dataset includes vehicle trajectory data (speed and position) and corresponding energy consumption records collected from a CAV platform equipped with lidar, cameras, onboard computation units, and communication modules. The primary objective is to quantify the baseline energy consumption attributable to sensing and computing systems, independent of any advanced cooperative control strategies. During experiments, the leading vehicle followed a predetermined velocity profile, and the following CAV mirrored this trajectory using a basic car-following control to ensure consistent driving behavior. This setup enables a reliable benchmark for assessing the energy cost introduced by onboard CDA hardware (e.g., lidar and GPU-based processing). The dataset is essential for evaluating energy baselines and supports future comparative studies involving additional cooperative strategies. ![system img](system.png) ![vector img](vector.png)

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Locations, metadata, and species cover from field sampling survey associated with NEON AOP survey, East River, CO 2018

Locations and descriptions of the sites where field sampling was conducted during the 2018 National Ecological Observatory Network (NEON) Airborne Observation Platform (AOP) imaging spectroscopy and lidar surveys in Gunnison County, Colorado. The sampling sites were located across East River, Washington Gulch, Slate River, and Coal Creek watersheds and contained a mixture of meadow, shrub, and tree sampling sites. This data package contains the location, metadata, species composition, and vector files for the sampling sites in this project. Species vouchers that have been preserved in the Rocky Mountain Biological Laboratory herbarium can be found on the Consortia of Southern Rocky Mountain Herbaria website: http://soroherbaria.org/portal/collections/index.php. Vector files provided here are for areas delineated based on the AOP reflectance data and mosaic developed by Brodrick et al. (doi: 10.15485/1618131), and use with other datasets should be closely evaluated. Subsequent data packages will contain additional biogeochemical, microbial, and geophysical data as they become available for these sampling sites. For full documentation, please see associated reference and additional references will be added here as they become available.

2018 NEON and 2025 CHESS Campaigns↗

Depth-resolved sagebrush root metabolomics, rhizosphere microbial communities, and geochemistry at the East River Watershed

This data set consists of results from soil nutrient profile, untargeted metabolomics, mass spec imaging, and amplicon sequencing. Data for soil nutrient profile includes common cations (Ca, Mg, Na, and K etc.) extracted from 3 digesting steps – ammonia acetate (for exchangeable cations), nitric acid (for acid dissolved fraction), and hydrofluoric acid/perchloric acid (HF/HClO4) for whole soil digestion. It also includes concentration of organic carbon, inorganic nitrogen (ammonia and nitrate) and phosphorus (Bray-1 P and nitric acid extract), and total nitrogen and phosphorus. Data for untargeted metabolomics includes metabolomic profile for root exudate/tissues and soil extracts from depths at surface soil to saprolite, that were measured using gas chromatography – mass spectrometry (GC-MS), and liquid chromatography – tandem mass spectrometry (LC-MS/MS). Data for mass spec imaging includes spatial distribution of metabolites that were detected and annotated with Fourier transformation ion cyclotron resonance mass spectrometer (FTICR-MS). Data for amplicon sequencing includes the base paired 16S and ITS ribosomal RNA sequences from Miseq Illumina sequencing. All samples were collected from 2 sampling campaign October 2022 and June 2023. Collectively, these datasets enable a mechanistic evaluation of how nutrient acquisition, especially nitrogen and phosphorus, differs between shallow roots operating in soil and deep roots functioning within the fractured bedrock zone. All files are provided as comma-separated values (CSV) fies (.csv) and (GZIP) file (.gz). The compressed .gz FASTQ files can be read directly in R using the dada2 package as part of the amplicon sequence analysis workflow. This work was supported by the Watershed Function Science Focus Area at Lawrence Berkeley National Laboratory funded by the US Department of Energy, Office of Science, Biological and Environmental Research under Contract No. DE-AC02-05CH11231. This research was performed on a project award 60563 (https://dx.doi.org/10.46936/expl.proj.2022.60563/60008727) from the Environmental Molecular Sciences Laboratory, a DOE Office of Science User Facility sponsored by the Biological and Environmental Research program under Contract No. DE-AC05-76RL01830.

EARTH SCIENCE > AGRICULTURE > SOILS > CARBON↗

Towards Query-Efficient Black-Box Adversary with Zeroth-Order Natural Gradient Descent

Despite the great achievements of the modern deep neural networks (DNNs), the vulnerability/robustness of state-of-the-art DNNs raises security concerns in many application domains requiring high reliability. Various adversarial attacks are proposed to sabotage the learning performance of DNN models. Among those, the black-box adversarial attack methods have received special attentions owing to their practicality and simplicity. Black-box attacks usually prefer less queries in order to maintain stealthy and low costs. However, most of the current black-box attack methods adopt the first-order gradient descent method, which may come with certain deficiencies such as relatively slow convergence and high sensitivity to hyper-parameter settings. In this paper, we propose a zeroth-order natural gradient descent (ZO-NGD) method to design the adversarial attacks, which incorporates the zeroth-order gradient estimation technique catering to the black-box attack scenario and the second-order natural gradient descent to achieve higher query efficiency. The empirical evaluations on image classification datasets demonstrate that ZO-NGD can obtain significantly lower model query complexities compared with state-of-the-art attack methods.

Zhao, Pu↗

CIAM Climatological Isolation and Attraction Model–Climatological Lagrangian Coherent Structures

Matlab code to compute climatological Lagrangian Coherent Structures. Climatological Lagrangian Coherent Structures have been shown to be effective in identifying regions that persistently attract or isolate trajectories at the sea surface, while also identifying recurrent transport patterns and transport barriers. It is therefore known as the Climatological Isolation and Attraction Model within NETL DoE. This repository also includes the information necessary to replicate the cCLS published in the paper which is freely available at: https://www.nature.com/articles/s41598-018-23121-y The method is developed in: Duran, R., F. J. Beron-Vera, M. J. Olascoaga (2018). Extracting quasi-steady Lagrangian transport patterns from the ocean circulation: An application to the Gulf of Mexico. Scientific Reports, 8(1), 5218. https://www.nature.com/articles/s41598-018-23121-y The method is successfully evaluated with a large dataset of satellite-tracked drifters, and with a different ocean model in: Gough M.K., F. J. Beron-Vera, M. J. Olascoaga, J. Sheinbaum, J. Jouenno, R. Duran (2019). Persistent Lagrangian transport patterns in the northwestern Gulf of Mexico. J. Phys. Oceanogr., 49, 353–367, https://doi.org/10.1175/JPO-D-17-0207.1 and with floats and yet a different model in: Maslo, A., Azevedo Correia de Souza, J. M., Andrade-Canto, F., & Rodríguez Outerelo, J. (2020). Connectivity of deep waters in the Gulf of Mexico. Journal of Marine Systems, 203, 103267. https://doi.org/10.1016/j.jmarsys.2019.103267

CIAM↗

JetNet: A Python package for accessing open datasets and benchmarking machine learning methods in high energy physics

JetNet is a Python package that aims to increase accessibility and reproducibility for machinelearning (ML) research in high energy physics (HEP), primarily related to particle jets. Basedon the popular PyTorch ML framework, it provides easy-to-access and standardized interfacesfor multiple heterogeneous HEP datasets and implementations of evaluation metrics, lossfunctions, and more general utilities relevant to HEP.

97 MATHEMATICS AND COMPUTING↗

Evaluation of Methods for Comparison of Spatiotemporal and Time Series Datasets

Data used by the National Renewable Energy Laboratory (NREL) in energy analysis are often produced by industry and licensed or purchased for analysis. While this practice provides needed flexibility in selecting data for analysis it presents challenges in understanding the differences among multiple, ostensibly similar, datasets. As options for source data become more varied, it is important to be able to articulate why certain datasets were chosen and to ensure those include the data that best meet the boundaries and/or limitations of a particular analysis. This report represents the first of three phases of research intended to develop methods to quantitatively assess and compare both input datasets and the results of analyses performed at NREL. This capability is critical to identifying tipping points in the costs or benefits of achieving high spatial and temporal resolution of input data.

97 MATHEMATICS AND COMPUTING↗

A library of AI-assisted FAIR water cycle and related disturbance datasets to enable model training, parameterization and validation

This whitepaper is responsive to focal area Data acquisition and assimilation enabled by machine learning, AI, and advanced methods. Here we describe how FAIR (Findable, Accessible, Reusable, Interoperable) datasets related to water cycle extremes are essential for successful implementation of ML in Earth System and other models. We also describe how AI can be used to acquire and integrate water cycle data related to extreme events to create a library of FAIR datasets for training and evaluating algorithms.

58 GEOSCIENCES↗

Seismic Signal Detection on International Monitoring System 3-Component Stations using PhaseNet

In this report we discuss training a deep learning seismic signal detection model on 3-component stations from the International Monitoring System (IMS) using the PhaseNet architecture. Using 14 years of associated signals from the International Data Centre’s (IDC) Late Event Bulletin (LEB), we auto-curated training data consisting of signal windows containing associated arrivals, and noise windows that contain no LEB-associated signals. We trained several models using different waveform window durations (30 seconds and 100 seconds), with and without bandpass filtering. We evaluated the effectiveness of our models using associated signals from the Unconstrained Global Event Bulletin (UGEB) and found that several of our models outperformed the signal detections from the IDC’s Selected Event List 3 (SEL3) arrival table. The SEL3 bulletin evaluated on the UGEB dataset with 100-second waveform windows registered a precision and recall of .15 and .48, respectively, versus .19 and .59 for our filtered-data model. For the 30-second waveform window dataset, the SEL3 bulletin achieved a precision and recall of .31 and .47, respectively, versus .32 and .60 for our filtered-data model. Finally, our models detected signals from all source-to-receiver distances, suggesting it is feasible to use a single PhaseNet model for the IMS network.

58 GEOSCIENCES↗