Search NASA⌕ Search

SEARCH · Search NASA

Results for “Labeled 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 397 records · Page 22

Text Mining for Process–Structure–Properties Relationships in Metals

With the advent of large language models (LLMs), the vast unstructured text within millions of academic papers is increasingly accessible for materials discovery—although significant challenges remain. While LLMs offer promising few- and zero-shot learning capabilities, particularly valuable in the materials domain where expert annotations are scarce, general-purpose LLMs often fail to address key materials-specific queries without further adaptation. To bridge this gap, fine-tuning LLMs on human-labeled data is essential for effective structured knowledge extraction (Liu in The Importance of Human-Labeled Data in the Era of LLMs, 2023). Here, in this study, we introduce a novel annotation schema designed to extract generic process–structure–properties relationships from scientific literature. We demonstrate the utility of this approach using a dataset of 128 abstracts, with annotations drawn from two distinct domains: high-temperature materials (Domain I) and uncertainty quantification in simulating materials microstructure (Domain II). Initially, we developed a conditional random field (CRF) model based on MatBERT—a domain-specific BERT variant—and evaluated its performance on Domain I. Subsequently, we compared this model with a fine-tuned LLM (GPT-4o from OpenAI) under identical conditions. Our results indicate that fine-tuning LLMs can significantly improve entity extraction performance over the BERT-CRF baseline on Domain I. However, when additional examples from Domain II were incorporated, the performance of the BERT-CRF model became comparable to that of the GPT-4o model. These findings underscore the potential of our schema for structured knowledge extraction and highlight the complementary strengths of both modeling approaches.

Materials science↗

An Efficient Bayesian Method for Advancing the Application of Deep Learning in Earth Science

Despite their wide and successful applications, deep learning (DL) models are prone to overfitting for small training datasets, produce a poor predictive performance for uncertain data, and provide point estimations without any indication of the accuracy and credibility. These limitations of the deterministic DL models hinder their effective application in Earth system science where the labelled data are sparse, noisy and incomplete with large uncertainty and where the predictive uncertainty quantification is needed for scientific understanding and policy decision making. Integration of Bayesian inference into DL models adds an estimate of uncertainty and regularization in the predictions. However, traditional Bayesian methods are computationally unaffordable and inflexible for high-dimensional problems, which limits their application to DL systems that typically have millions of model parameters. In this effort, we propose an efficient and general-purpose Bayesian inference method to advance DL model optimization and uncertainty quantification, so as to facilitate the adoption of DL in Earth sciences. In a demonstration, we integrate the proposed Bayesian method with a feedforward neural network (NN) to build a fast-to-evaluate surrogate of the complex Energy Exascale Earth System Land Model for efficient modeling. The formulated Bayesian NN, using a small number of training data, produces an accurate prediction with high credibility, whereas with the same small training size, the deterministic NN cannot yield a reasonable estimation and does not provide confidence information. The proposed Bayesian method is computationally efficient and flexible, capable of integration with diverse network variants such as convolutional NNs and recurrent NNs to advance the application of DL in Earth sciences.

Lu, Dan↗

TH-CD-206-01: Expectation-Maximization Algorithm-Based Tissue Mixture Quantification for Perfusion MRI

Purpose: To investigate the feasibility of estimating the tissue mixture perfusions and quantifying cerebral blood flow change in arterial spin labeled (ASL) perfusion MR images. Methods: The proposed perfusion MR image analysis framework consists of 5 steps: (1) Inhomogeneity correction was performed on the T1- and T2-weighted images, which are available for each studied perfusion MR dataset. (2) We used the publicly available FSL toolbox to strip off the non-brain structures from the T1- and T2-weighted MR images. (3) We applied a multi-spectral tissue-mixture segmentation algorithm on both T1- and T2-structural MR images to roughly estimate the fraction of each tissue type - white matter, grey matter and cerebral spinal fluid inside each image voxel. (4) The distributions of the three tissue types or tissue mixture across the structural image array are down-sampled and mapped onto the ASL voxel array via a co-registration operation. (5) The presented 4-dimensional expectation-maximization (4D-EM) algorithm takes the down-sampled three tissue type distributions on perfusion image data to generate the perfusion mean, variance and percentage images for each tissue type of interest. Results: Experimental results on three volunteer datasets demonstrated that the multi-spectral tissue-mixture segmentation algorithm was effective to initialize tissue mixtures from T1- and T2-weighted MR images. Compared with the conventional ASL image processing toolbox, the proposed 4D-EM algorithm not only generated comparable perfusion mean images, but also produced perfusion variance and percentage images, which the ASL toolbox cannot obtain. It is observed that the perfusion contribution percentages may not be the same as the corresponding tissue mixture volume fractions estimated in the structural images. Conclusion: A specific application to brain ASL images showed that the presented perfusion image analysis method is promising for detecting subtle changes in tissue perfusions, which is valuable for the early diagnosis of certain brain diseases, e.g. multiple sclerosis.

60 APPLIED LIFE SCIENCES↗

LCLS RF Station Anomaly Candidates, SLAC, Nov 2020 to Dec 2020, Version 1

The data consists of radio-frequency (RF) station process variables (amplitude and amplitude-mean-out-of-tolerance bit) and beam position monitor (BPM) values for about a month of Linac Coherent Light Source (LCLS) time. The data is all from normal conducting hard X-ray operation. The dataset contains roughly 5000 anomaly “candidates.” For each candidate, we include 3.5 minutes of data from the 82 RF stations and ~20 seconds of data from 4 BPMs. We also include several CSV files with additional information and labels for each candidate.

43 PARTICLE ACCELERATORS↗

Optical Wavelength Guided Self-Supervised Feature Learning For Galaxy Cluster Richness Estimate

Most galaxies in the nearby Universe are gravitationally bound to a cluster or group of galaxies. Their optical contents, such as optical richness, are crucial for understanding the co-evolution of galaxies and large-scale structures in modern astronomy and cosmology. The determination of optical richness can be challenging. We propose a self-supervised approach for estimating optical richness from multi-band optical images. The method uses the data properties of the multi-band optical images for pre-training, which enables learning feature representations from a large but unlabeled dataset. We apply the proposed method to the Sloan Digital Sky Survey. The result shows our estimate of optical richness lowers the mean absolute error and intrinsic scatter by 11.84% and 20.78%, respectively, while reducing the need for labeled training data by up to 60%. We believe the proposed method will benefit astronomy and cosmology, where a large number of unlabeled multi-band images are available, but acquiring image labels is costly.

79 ASTRONOMY AND ASTROPHYSICS↗

An Open-Access Repository of Synchrophasor Data Quality Examples: Curation and Example Applications

Synchrophasor measurements are critical in providing wide-area situational awareness to power system operators. However, data artifacts may be introduced due to various issues such as loss of communication, loss of GPS signal, internal clock error, and vendor-specific implementation of phasor estimation algorithms. Tools designed to provide actionable insights from synchrophasor data, hence, must be designed to be robust to these data quality issues. In this work, two years of synchrophasor data sourced from multiple electric utilities in the United States were analyzed to identify examples of data quality problems. These examples were then labeled and published in the Grid Event Signature Library, a publicly available repository of power system measurements hosted by the Oak Ridge National Laboratory. This paper describes the data curation process, and illustrates two application use cases where the dataset can be valuable to the research community. In the first use case, a random forest classifier is trained to distinguish power system disturbance signatures from data anomalies introduced in synchrophasor measurements due to clock errors. The second use case studies the impact of data quality issues on an example synchrophasor application (specifically, event start time determination). The choice of data quality problems investigated is informed by the examples in the repository curated in this work.

24 POWER TRANSMISSION AND DISTRIBUTION↗

PMU Data Quality and Sensor Health Monitoring

Phasor Measurement Units (PMUs) play a critical role in the evolution of the electric power industry by providing high-precision, real-time monitoring of essential power system metrics. However, effectively detecting abnormalities and critical events from PMU data is a complex task, complicated by intricate temporal patterns, a scarcity of labeled data for training algo- rithms, and constraints on online computational power. In this study, we apply TranAD, an innovative algorithm that combines transformer architectures with the refinement of adversarial learning, to both synthetic and real-world PMU datasets for developing a data quality and sensor online health monitoring platform for utilities. Our findings reveal that TranAD not only provides efficient detection and localization but also enhances the detail with which abnormalities are detected, marking a a significant step forward in the field of clean data acquisition processes for power system monitoring

deep neural network, machine learning (ML)↗

Fine-tuning TrailMap: The utility of transfer learning to improve the performance of deep learning in axon segmentation of light-sheet microscopy images

Light-sheet microscopy has made possible the 3D imaging of both fixed and live biological tissue, with samples as large as the entire mouse brain. However, segmentation and quantification of that data remains a time-consuming manual undertaking. Machine learning methods promise the possibility of automating this process. This study seeks to advance the performance of prior models through optimizing transfer learning. We fine-tuned the existing TrailMap model using expert-labeled data from noradrenergic axonal structures in the mouse brain. By changing the cross-entropy weights and using augmentation, we demonstrate a generally improved adjusted F1-score over using the originally trained TrailMap model within our test datasets.

97 MATHEMATICS AND COMPUTING↗

C fluxes from incubations of California upland soil with and without synthetic iron (hydr)oxide associations

There is a growing understanding that interactions with soil mineral phases contribute to the accumulation and retention of otherwise degradable organic matter in soils and sediments. However, the bioavailability of organic compounds in mineral-organic-associations, especially under varying environmental conditions is not well known. To assess the impact of mineral association and warming on the decomposition of an easily respirable organic substrate (glucose), we conducted a series of laboratory incubations at different temperatures with field-collected soils from 10 to 20 cm, 50–60 cm, and 80–90 cm depth at the Blodgett Forest Research Station. We added 13C-labeled glucose either directly to native soil or sorbed to one of two synthetic iron (hydr)oxide phases (goethite and ferrihydrite) that differ in crystallinity and affinity for sorbing glucose. We incubated the soils at 25 deg C. This dataset comprises the added glucose C and the cumulative fluxes over 72 and 1896 hours. We found that: (1) association with the Fe (hydr)oxide minerals reduced the decomposition rate of glucose by > 99.5% relative to rate of decomposition for free glucose in soil; (2) the respiration rate per gram carbon did not differ appreciably with depth, suggesting a similar degree of decomposability for native C across depths and that under the incubation conditions total carbon availability represents the principal limitation on respiration under these conditions as opposed to reduced abundance of decomposers or moisture and oxygen limitations; (3) addition of free glucose enhanced native carbon respiration at all soil depths with the largest effect at 50–60 cm; (4) in general respiration of the organo-mineral complex (glucose and iron-(hydr)oxide) was less temperature sensitive than was respiration of native carbon; (5) the addition of organic free mineral decreased the rate of soil respiration in the intermediate 50–60 cm depth soil. The results emphasize the key role of MOAs in regulating the fluxes of carbon from soils to the atmosphere and in turn the stocks of soil carbon.This dataset lists the glucose additions made to soils with and without added geothite or ferrihidrite and the C respired over the timespan of the incubations.

54 ENVIRONMENTAL SCIENCES↗

Laboratory time series moisture manipulative experiment from sediment across San Antonio, Texas: time series aerobic respiration and geochemistry

This dataset supports a broader study examining the effects of wetting and drying on hyporheic zone respiration. The dataset provides data generated from a laboratory moisture manipulation experiment. The contents include time series aerobic respiration and moisture; dissolved oxygen; sediment geochemistry data; and field metadata (including qualitative information on instream and river corridor characteristics). Samples were collected as part of the WHONDRS Allison Veach collaboration (AV1). The data package associated with the AV1 study is available at https://data.ess-dive.lbl.gov/datasets/doi:10.15485/2529428. AV1 sampling occurred across 7 perennial and 7 intermittent streams in San Antonio, Texas. Each stream/site was visited both in summer during base flow (July-September 2023) and winter during peak flow (January-February 2024). This study uses subsamples from a subset of AV1 samples. The original field samples were labeled as AV1_###. Subsequent subsamples for this study were labeled as EV_###. The labels from the field samples and the EV subsamples can be mapped directly based on the digits following the prefix and underscore (i.e., EV_001 is a subsample from AV1_001). See the critical details section below for more details on sample naming. For details on how to navigate data packages generated by this project, see https://data.ess-dive.lbl.gov/portals/PNNLRiverCorridorSFA/About. This dataset is comprised of one main data folder containing (1) file-level metadata; (2) data dictionary; (3) field metadata; (4) readme; (5) field protocol; and a (6) a subfolder with sediment sample data from the incubation experiment. The sample data subfolder contains (1) effect size; (2) iron (II); (3) gravimetric moisture; (4) respiration rates; (5) raw dissolved oxygen values and plots; (6) specific conductance; (7) pH; (8) temperature; (9) a summary containing mean, median, and standard deviation values of each data type for each treatment (wet and dry); and (10) methods codes. All files are .csv or.pdf.

54 ENVIRONMENTAL SCIENCES↗

Probing biolabels for high throughput biosensing via synchrotron radiation SEIRA technique

Bio-diagnostic assays of high complexity rely on nanoscaled assay recognition elements that can provide unique selectivity and design-enhanced sensitivity features. High throughput performance requires the simultaneous detection of various analytes combined with appropriate bioassay components. Nanoparticle induced sensitivity enhancement, and subsequent multiplexed capability Surface-Enhanced InfraRed Absorption (SEIRA) assay formats are fitting well these purposes. SEIRA constitutes an ideal platform to isolate the vibrational signatures of targeted bioassay and active molecules. The potential of several targeted biolabels, here fluorophore-labeled antibody conjugates, chemisorbed onto low-cost biocompatible gold nano-aggregates substrates have been explored for their use in assay platforms. Dried films were analyzed by synchrotron radiation based FTIR/SEIRA spectro-microscopy and the resulting complex hyperspectral datasets were submitted to automated statistical analysis, namely Principal Components Analysis (PCA). The relationships between molecular fingerprints were put in evidence to highlight their spectral discrimination capabilities. We demonstrate that robust spectral encoding via SEIRA fingerprints opens up new opportunities for fast, reliable and multiplexed high-end screening not only in biodiagnostics but also in vitro biochemical imaging.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

AI Model Benchmarking for Nonproliferation Applications: Steel Thread Benchmarking Task Force Technical Report (Rev. 2)

Steel Thread is a NA-22 venture that seeks to build trustworthy, reliable AI models that can be used in a wide variety of nonproliferation tasks. A key aspect of building these models is developing appropriate benchmarks and evaluation methods, which will enable the venture to identify and adapt models to provide the most value in the nonproliferation domain. Benchmarks must be relevant to key tasks in this domain, such as question answering, information retrieval, document summarization and classification, consensus analysis, and image and data analysis. This report 1) provides an overview of benchmark design, evaluation, and challenges; 2) reviews a variety of open benchmarks, with a focus on language models and tasks; and 3) identifies benchmarks that are most relevant to Steel Thread. This report is intended to serve as a basis for further efforts to classify and evaluate benchmarks and their correlation with success on nonproliferation-specific tasks. The Steel Thread venture has defined benchmarks to be a particular combination of a dataset (or datasets) and a metric (or metrics) conceptualized as representing one or more specific tasks or sets of abilities for a specific modality. It is adopted by a research community as a shared framework for comparing methods.1 It includes 1) Data: Labeled (a designated subset not used for training, which could be all the data), 2) Metric: A way to quantify performance, 3) Task/Ability: What the benchmark is testing, 4) Protocol: A structured and repeatable evaluation process, 5) Baseline/Reference Model: For comparison; could be statistical, rule-based, SME-derived, or another model, and 6) Maintenance Plan: to update with new information over time; important for long-term utility. For further clarity, the definition includes what a benchmark, in this context, is not. It is not a corpus of training data, specific to a model (it is intended to apply to a range of models), a universal evaluation of performance, a guarantee that the ‘top’ model on the leaderboard will be the best fit for every specific use case, an all-encompassing proof of a model’s universal quality, nor is it a one-size-fits-all measure of success. It does not cover every real-world constraint (like operational, ethical, or cost considerations), a systems integration test, or a unit test. This definition was inspired by and resulted from discussions within the Steel Thread Benchmarking Task Force. This group was formed to define what we would mean as a benchmark within Steel Thread but persisted as the need to develop a thorough understanding of the large and expanding existing benchmarking space. This technical report is a result of the group’s divide and conquer approach to exploring this space. The release of benchmarks might not be progressing as quickly as model development, but it is moving very fast, as many benchmarks quickly become saturated, when state-of-the-art models score so close to the benchmark’s ceiling that their results are virtually indistinguishable. At that point, the test no longer differentiates between new systems, so researchers usually stop reporting scores as the benchmark no longer informs about improvements from the next generation of models. In the OpenAI announcement of GPT-5, they reported results on six flagship public benchmarks (AIME 2025, SWE-bench Verified, Aider Polyglot, MMMU, HealthBench Hard, GPQA) but the full system-card covers roughly thirty-five separate evaluations, comprising hundreds of test task items in total. There have been some efforts to summarize benchmarks in specific fields, like for text-to-image generation, but these surveys have had a narrow methodology scope. Therefore, a comprehensive survey of all benchmarks or even all benchmarks that could be relevant to Steel Thread is outside of the scope of this report. We chose some specific benchmarks to investigate in detail.

97 MATHEMATICS AND COMPUTING↗

Automated detection and analysis of particle beams in laser-plasma accelerator simulations

Numerical simulations of laser-plasma wakefield (particle) accelerators model the acceleration of electrons trapped in plasma oscillations (wakes) left behind when an intense laser pulse propagates through the plasma. The goal of these simulations is to better understand the process involved in plasma wake generation and how electrons are trapped and accelerated by the wake. Understanding of such accelerators, and their development, offer high accelerating gradients, potentially reducing size and cost of new accelerators. One operating regime of interest is where a trapped subset of electrons loads the wake and forms an isolated group of accelerated particles with low spread in momentum and position, desirable characteristics for many applications. The electrons trapped in the wake may be accelerated to high energies, the plasma gradient in the wake reaching up to a gigaelectronvolt per centimeter. High-energy electron accelerators power intense X-ray radiation to terahertz sources, and are used in many applications including medical radiotherapy and imaging. To extract information from the simulation about the quality of the beam, a typical approach is to examine plots of the entire dataset, visually determining the adequate parameters necessary to select a subset of particles, which is then further analyzed. This procedure requires laborious examination of massive data sets over many time steps using several plots, a routine that is unfeasible for large data collections. Demand for automated analysis is growing along with the volume and size of simulations. Current 2D LWFA simulation datasets are typically between 1GB and 100GB in size, but simulations in 3D are of the order of TBs. The increase in the number of datasets and dataset sizes leads to a need for automatic routines to recognize particle patterns as particle bunches (beam of electrons) for subsequent analysis. Because of the growth in dataset size, the application of machine learning techniques for scientific data mining is increasingly considered. In plasma simulations, Bagherjeiran et al. presented a comprehensive report on applying graph-based techniques for orbit classification. They used the KAM classifier to label points and components in single and multiple orbits. Love et al. conducted an image space analysis of coherent structures in plasma simulations. They used a number of segmentation and region-growing techniques to isolate regions of interest in orbit plots. Both approaches analyzed particle accelerator data, targeting the system dynamics in terms of particle orbits. However, they did not address particle dynamics as a function of time or inspected the behavior of bunches of particles. Ruebel et al. addressed the visual analysis of massive laser wakefield acceleration (LWFA) simulation data using interactive procedures to query the data. Sophisticated visualization tools were provided to inspect the data manually. Ruebel et al. have integrated these tools to the visualization and analysis system VisIt, in addition to utilizing efficient data management based on HDF5, H5Part, and the index/query tool FastBit. In Ruebel et al. proposed automatic beam path analysis using a suite of methods to classify particles in simulation data and to analyze their temporal evolution. To enable researchers to accurately define particle beams, the method computes a set of measures based on the path of particles relative to the distance of the particles to a beam. To achieve good performance, this framework uses an analysis pipeline designed to quickly reduce the amount of data that needs to be considered in the actual path distance computation. As part of this process, region-growing methods are utilized to detect particle bunches at single time steps. Efficient data reduction is essential to enable automated analysis of large data sets as described in the next section, where data reduction methods are steered to the particular requirements of our clustering analysis. Previously, we have described the application of a set of algorithms to automate the data analysis and classification of particle beams in the LWFA simulation data, identifying locations with high density of high energy particles. These algorithms detected high density locations (nodes) in each time step, i.e. maximum points on the particle distribution for only one spatial variable. Each node was correlated to a node in previous or later time steps by linking these nodes according to a pruned minimum spanning tree (PMST). We call the PMST representation 'a lifetime diagram', which is a graphical tool to show temporal information of high dense groups of particles in the longitudinal direction for the time series. Electron bunch compactness was described by another step of the processing, designed to partition each time step, using fuzzy clustering, into a fixed number of clusters.

43 PARTICLE ACCELERATORS↗

The Livermore Brain: Massive Deep Learning Networks Enabled by High Performance Computing

The proliferation of inexpensive sensor technologies like the ubiquitous digital image sensors has resulted in the collection and sharing of vast amounts of unsorted and unexploited raw data. Companies and governments who are able to collect and make sense of large datasets to help them make better decisions more rapidly will have a competitive advantage in the information era. Machine Learning technologies play a critical role for automating the data understanding process; however, to be maximally effective, useful intermediate representations of the data are required. These representations or “features” are transformations of the raw data into a form where patterns are more easily recognized. Recent breakthroughs in Deep Learning have made it possible to learn these features from large amounts of labeled data. The focus of this project is to develop and extend Deep Learning algorithms for learning features from vast amounts of unlabeled data and to develop the HPC neural network training platform to support the training of massive network models. This LDRD project succeeded in developing new unsupervised feature learning algorithms for images and video and created a scalable neural network training toolkit for HPC. Additionally, this LDRD helped create the world’s largest freely-available image and video dataset supporting open multimedia research and used this dataset for training our deep neural networks. This research helped LLNL capture several work-for-others (WFO) projects, attract new talent, and establish collaborations with leading academic and commercial partners. Finally, this project demonstrated the successful training of the largest unsupervised image neural network using HPC resources and helped establish LLNL leadership at the intersection of Machine Learning and HPC research.

97 MATHEMATICS AND COMPUTING↗

Empirical Analysis and Automated Classification of Security Bug Reports

With the ever expanding amount of sensitive data being placed into computer systems, the need for effective cybersecurity is of utmost importance. However, there is a shortage of detailed empirical studies of security vulnerabilities from which cybersecurity metrics and best practices could be determined. This thesis has two main research goals: (1) to explore the distribution and characteristics of security vulnerabilities based on the information provided in bug tracking systems and (2) to develop data analytics approaches for automatic classification of bug reports as security or non-security related. This work is based on using three NASA datasets as case studies. The empirical analysis showed that the majority of software vulnerabilities belong only to a small number of types. Addressing these types of vulnerabilities will consequently lead to cost efficient improvement of software security. Since this analysis requires labeling of each bug report in the bug tracking system, we explored using machine learning to automate the classification of each bug report as a security or non-security related (two-class classification), as well as each security related bug report as specific security type (multiclass classification). In addition to using supervised machine learning algorithms, a novel unsupervised machine learning approach is proposed. An ac- curacy of 92%, recall of 96%, precision of 92%, probability of false alarm of 4%, F-Score of 81% and G-Score of 90% were the best results achieved during two-class classification. Furthermore, an accuracy of 80%, recall of 80%, precision of 94%, and F-score of 85% were the best results achieved during multiclass classification.

Cybersecurity↗

Automated classification of scientific publications linked to GES DISC datasets

The data collections archived and distributedby the GES DISC NASA data center arewidely utilized for various Earth Science studies.As these collections are created, many researchworks are published regarding the collections, algorithms,validations and applications. SinceGES DISC collects these publications and providestheir citations for the users, it is helpful tocategorize them based on how they relate to the datasetsthey are associated with. Specifically,whether the publication that is linked to GES DISCdataset is using it for applicational research,or if it describes the algorithm for dataset creation,or the validation of the dataset, or providesthe general overview of the data collection. Currently,this process requires simple manuallabelling, and as such, may be possible to solve viaautomation. To approach this problem, wedeveloped machine learning classifiers to predictthe category a publication belongs to. We usedmanually labeled publications as training data forsupervised machine learning algorithms:Random Forest and Naive Bayes. We achieved classificationaccuracy that is substantially betterthan the baseline accuracy, thus greatly improvingthe efficiency of the publication internalanalysis.

Rohan Dayal↗

SIDDA: SInkhorn Dynamic Domain Adaptation for image classification with equivariant neural networks

Modern neural networks (NNs) often do not generalize well in the presence of a ‘covariate shift’; that is, in situations where the training and test data distributions differ, but the conditional distribution of classification labels given the data remains unchanged. In such cases, NN generalization can be reduced to a problem of learning more robust, domain-invariant features. Domain adaptation (DA) methods include a broad range of techniques aimed at achieving this; however, these methods have struggled with the need for extensive hyperparameter tuning, which then incurs significant computational costs. In this work, we introduce SInkhorn Dynamic Domain Adaptation (SIDDA), an out-of-the-box DA training algorithm built upon the Sinkhorn divergence, that can achieve effective domain alignment with minimal hyperparameter tuning and computational overhead. We demonstrate the efficacy of our method on multiple simulated and real datasets of varying complexity, including simple shapes, handwritten digits, real astronomical observations, and remote sensing data. These datasets exhibit covariate shifts due to noise, blurring, differences between telescopes, and variations in imaging wavelengths. SIDDA is compatible with a variety of NN architectures, and it works particularly well in improving classification accuracy and model calibration when paired with symmetry-aware equivariant NNs (ENNs). We find that SIDDA consistently enhances the generalization capabilities of NNs, achieving up to a ${\approx}40\%$ improvement in classification accuracy on unlabeled target data, while also providing a more modest performance gain of $\lesssim 1\%$ on labeled source data. We also study the efficacy of DA on ENNs with respect to the varying group orders of the dihedral group DN, and find that the model performance improves as the degree of equivariance increases. Finally, if SIDDA achieves proper domain alignment, it also enhances model calibration on both source and target data, with the most significant gains in the unlabeled target domain—achieving over an order of magnitude improvement in the expected calibration error and Brier score. SIDDA’s versatility across various NN models and datasets, combined with its automated approach to domain alignment, has the potential to significantly advance multi-dataset studies by enabling the development of highly generalizable models.

79 ASTRONOMY AND ASTROPHYSICS↗

Reduced volume of diabetic pancreatic islets in rodents detected by synchrotron X-ray phase-contrast microtomography and deep learning network

The pancreatic islet is a highly structured micro-organ that produces insulin in response to rising blood glucose. Here we develop a label-free and automatic imaging approach to visualize the islets in situ in diabetic rodents by the synchrotron radiation X-ray phase-contrast microtomography (SRμCT) at the ID17 station of the European Synchrotron Radiation Facility. The large-size images (3.2 mm × 15.97 mm) were acquired in the pancreas in STZ-treated mice and diabetic GK rats. Each pancreas was dissected by 3000 reconstructed images. The image datasets were further analysed by a self-developed deep learning method, AA-Net. All islets in the pancreas were segmented and visualized by the three-dimension (3D) reconstruction. After quantifying the volumes of the islets, we found that the number of larger islets (=>1500 μm 3 ) was reduced by 2-fold (wt 1004 ± 94 vs GK 419 ± 122, P < 0.001) in chronically developed diabetic GK rat, while in STZ-treated diabetic mouse the large islets were decreased by half (189 ± 33 vs 90 ± 29, P < 0.001) compared to the untreated mice. Our study provides a label-free tool for detecting and quantifying pancreatic islets in situ. It implies the possibility of monitoring the state of pancreatic islets in vivo diabetes without labelling.

59 BASIC BIOLOGICAL SCIENCES↗