Search NASA⌕ Search

SEARCH · Search NASA

Results for “LABELLING”

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 667 records · Page 37
All sourcesNASA NTRSDOE OSTI
NASA NTRS2011

Category Learning Research in the Interactive Online Environment Second Life

The interactive online environment Second Life allows users to create novel three-dimensional stimuli that can be manipulated in a meaningful yet controlled environment. These features suggest Second Life's utility as a powerful tool for investigating how people learn concepts for unfamiliar objects. The first of two studies was designed to establish that cognitive processes elicited in this virtual world are comparable to those tapped in conventional settings by attempting to replicate the established finding that category learning systematically influences perceived similarity . From the perspective of an avatar, participants navigated a course of unfamiliar three-dimensional stimuli and were trained to classify them into two labeled categories based on two visual features. Participants then gave similarity ratings for pairs of stimuli and their responses were compared to those of control participants who did not learn the categories. Results indicated significant compression, whereby objects classified together were judged to be more similar by learning than control participants, thus supporting the validity of using Second Life as a laboratory for studying human cognition. A second study used Second Life to test the novel hypothesis that effects of learning on perceived similarity do not depend on the presence of verbal labels for categories. We presented the same stimuli but participants classified them by selecting between two complex visual patterns designed to be extremely difficult to label. While learning was more challenging in this condition , those who did learn without labels showed a compression effect identical to that found in the first study using verbal labels. Together these studies establish that at least some forms of human learning in Second Life parallel learning in the actual world and thus open the door to future studies that will make greater use of the enriched variety of objects and interactions possible in simulated environments compared to traditional experimental situations.

Andrews, Jan↗
NASA NTRS2008

Multiple-Instance Regression with Structured Data

We present a multiple-instance regression algorithm that models internal bag structure to identify the items most relevant to the bag labels. Multiple-instance regression (MIR) operates on a set of bags with real-valued labels, each containing a set of unlabeled items, in which the relevance of each item to its bag label is unknown. The goal is to predict the labels of new bags from their contents. Unlike previous MIR methods, MI-ClusterRegress can operate on bags that are structured in that they contain items drawn from a number of distinct (but unknown) distributions. MI-ClusterRegress simultaneously learns a model of the bag's internal structure, the relevance of each item, and a regression model that accurately predicts labels for new bags. We evaluated this approach on the challenging MIR problem of crop yield prediction from remote sensing data. MI-ClusterRegress provided predictions that were more accurate than those obtained with non-multiple-instance approaches or MIR methods that do not model the bag structure.

learning↗
NASA NTRS2024

Automatic Detection and Classification of Aurora in THEMIS All‐Sky Images

We report a novel machine-learning algorithm for automatically detecting and classifying aurora in all–sky images (ASI) that is largely trained without requiring ground–truth labels. By including a small number of labeled images, we are able to automatically label all of the approximately 700 million images in the Time History of Events and Macroscale Interactions during Substorms (THEMIS) ASI data set from 2008 to 2022. We use a two–stage approach. In the first stage, we adapt the Simple framework for Contrastive Learning of Representations (SimCLR) algorithm to learn latent representations of THEMIS all–sky images. We then finetune a classifier network on the latent representations our model learns of the manually labeled Oslo aurora THEMIS (OATH) data set. We demonstrate that this two–stage approach achieves excellent classification results on data for which there is no current ML classification benchmark. The outcome of this work will facilitate efficient information retrieval for researchers interested in specific categories of aurora and will enable large scale statistical studies and machine learning analyses of THEMIS all–sky images that have not previously been possible. To demonstrate possible ways to utilize this database, we performed a statistical analysis of the occurrence rates of auroral labels with respect to solar wind parameters, interplanetary magnetic field vector, and geomagnetic indices. We further investigate the occurrence rates of auroral phenomena in the annotated data set and their geoeffectiveness by utilizing the co–located THEMIS ground magnetometer data set.

Jeremiah W Johnson↗
DOE OSTI2022

Generative memory for lifelong machine learning

Techniques are disclosed for training machine learning systems. An input device receives training data comprising pairs of training inputs and training labels. A generative memory assigns training inputs to each archetype task of a plurality of archetype tasks, each archetype task representative of a cluster of related tasks within a task space and assigns a skill to each archetype task. The generative memory generates, from each archetype task, auxiliary data comprising pairs of auxiliary inputs and auxiliary labels. A machine learning system trains a machine learning model to apply a skill assigned to an archetype task to training and auxiliary inputs assigned to the archetype task to obtain output labels corresponding to the training and auxiliary labels associated with the training and auxiliary inputs assigned to the archetype task to enable scalable learning to obtain labels for new tasks for which the machine learning model has not previously been trained.

Nadamuni Raghavan, Aswin↗
DOE OSTI2024

Predicting Drug Effects from High-dimensional Asymmetric Drug Data Sets using Graph Neural Networks: A Comprehensive Analysis of Multi-target Drug Effect Prediction

Graph neural networks (GNNs) have emerged as one of the most effective Machine learning (ML) techniques for drug effect prediction from drug molecular graphs. Despite having immense potential, GNN models lack performance when using data sets that contain high dimensional asymmetrically co-occurrent drug effects as targets with complex correlations between them. Training individual learning models for each drug effect and incorporating every prediction result for a wide spectrum of drug effects is beyond practicality. Such an implication provides a testbed to address this challenge as multi-target prediction problems, aiming to predict all drug effects at a time. We develop standard and hybrid graph neural networks (GNNs)to perform two separate tasks that are multi-regression for continuous values and multi-label classification for categorical values contained in our data sets. Since this step makes the target data even more sparse and introduces asymmetric label co-occurrence, the learning of multi-label classification models becomes difficult and heavily impacts the GNN's performance. To address these challenges, we propose a new data oversampling technique to improve multi-label classification performances on all the given imbalanced molecular graph data sets. Using the technique, we improve the data imbalance ratio of the drug effects better than before while protecting the data set's integrity. Finally, we evaluate multi-label classification performance using the best-performant hybrid GNN model on all the oversampled data sets obtained from the proposed oversampling technique. These results outperform those of other ML models including GNN models when they are trained on the original data sets or oversampled data sets using MLSMOTE (a well-known oversampling technique) in all evaluation metrics precision, recall, and F1 score by a significant margin.

Bose, Avishek [ORNL]↗
DOE OSTI2021

The Impact of Variability in Patient Exposure During Premarket Clinical Development on Postmarket Safety Outcomes

We characterized the size of the premarket safety population for 278 small-molecule new molecular entities (NMEs) and 61 new therapeutic biologics (NTBs) approved by the US Food and Drug Administration (FDA) between October 1, 2002, and December 31, 2014, evaluating the relationship of premarket safety population size to regulatory characteristics and postmarket safety outcomes. The median size of the safety population was 1,044, and was lower for NTBs than NMEs (median: 920 vs. 1,138, P = 0.04), orphan products than nonorphan products (393 vs. 1,606, P < 0.001), and for products with fast-track designation (617 vs. 1,455, P < 0.001), priority review (630 vs. 1,735, P < 0.001), and accelerated approval (475 vs. 1,164, P < 0.001), than products without that designation. The median number of postmarket safety label updates and issues added to the label were higher with larger premarket exposure among nonorphan products, but not among orphan products. Products with accelerated approval using a surrogate end point had a higher median number of safety issues added to the label than those with full approval, but this did not vary with the size of the safety population; fast-track and priority review were not associated with the number of safety issues added to the label. Here, a smaller safety population size was associated with a longer time to first safety outcome for nonorphan products but not orphan products. For orphan and nonorphan products combined, smaller premarket safety population size is not associated with the number or timing of postmarket safety outcomes, regardless of expedited program participation.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

Root identification in minirhizotron imagery with multiple instance learning

In this study, multiple instance learning (MIL) algorithms to automatically perform root detection and segmentation in minirhizotron imagery using only image-level labels are proposed. Root and soil characteristics vary from location to location, and thus, supervised machine learning approaches that are trained with local data provide the best ability to identify and segment roots in minirhizotron imagery. However, labeling roots for training data (or otherwise) is an extremely tedious and time-consuming task. This paper aims to address this problem by labeling data at the image level (rather than the individual root or root pixel level) and train algorithms to perform individual root pixel level segmentation using MIL strategies. Three MIL methods (multiple instance adaptive cosine coherence estimator, multiple instance support vector machine, multiple instance learning with randomized trees) were applied to root detection and compared to non-MIL approaches. The results show that MIL methods improve root segmentation in challenging minirhizotron imagery and reduce the labeling burden. In our results, multiple instance support vector machine outperformed other methods. The multiple instance adaptive cosine coherence estimator algorithm was a close second with an added advantage that it learned an interpretable root signature which identified the traits used to distinguish roots from soil and did not require parameter selection.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Dense autoencoders, clustering techniques, and semi-supervised learning for HPGe $γ$-spectra

Classifying high-resolution gamma spectra by their isotopic content is an essential task in nuclear forensics and other applications. Traditional analysis methods are often time-intensive, but machine learning (ML) may help analysts quickly process many spectra. Such methods tend to rely on abundant, well-labeled data for training. Historical gamma data exists in various fields but is not uniformly useful for supervised ML due to inconsistent labeling. Here, to address some of these challenges, we present a method to classify and organize unlabeled data from high-purity germanium detectors using an autoencoding neural network (autoencoder). We trained dense autoencoders to compress gamma data into latent representations that enable efficient data characterization. By clustering the encoded spectra or lower-dimensional mappings of them, we identified and removed portions of over-abundant data categories, resulting in a more balanced dataset and improved autoencoder performance. This encoding and clustering pipeline also enabled the organization of spectra into self-consistent categories. Finally, we found that encoded representations showed potential as inputs for semi-supervised learning of nuclide identification (NID) labels, achieving an average F1 score of 0.85 ± 0.03 when mapping encodings to a set of 65 isotope labels.

Autoencoders↗
DOE OSTI2016

3D-SIFT-Flow for atlas-based CT liver image segmentation

Purpose: In this paper, the authors proposed a new 3D registration algorithm, 3D-scale invariant feature transform (SIFT)-Flow, for multiatlas-based liver segmentation in computed tomography (CT) images. Methods: In the registration work, the authors developed a new registration method that takes advantage of dense correspondence using the informative and robust SIFT feature. The authors computed the dense SIFT features for the source image and the target image and designed an objective function to obtain the correspondence between these two images. Labeling of the source image was then mapped to the target image according to the former correspondence, resulting in accurate segmentation. In the fusion work, the 2D-based nonparametric label transfer method was extended to 3D for fusing the registered 3D atlases. Results: Compared with existing registration algorithms, 3D-SIFT-Flow has its particular advantage in matching anatomical structures (such as the liver) that observe large variation/deformation. The authors observed consistent improvement over widely adopted state-of-the-art registration methods such as ELASTIX, ANTS, and multiatlas fusion methods such as joint label fusion. Experimental results of liver segmentation on the MICCAI 2007 Grand Challenge are encouraging, e.g., Dice overlap ratio 96.27% ± 0.96% by our method compared with the previous state-of-the-art result of 94.90% ± 2.86%. Conclusions: Experimental results show that 3D-SIFT-Flow is robust for segmenting the liver from CT images, which has large tissue deformation and blurry boundary, and 3D label transfer is effective and efficient for improving the registration accuracy.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

Scale-Free Compact Routing Schemes in Networks of Low Doubling Dimension

In this work, we consider compact routing schemes in networks of low doubling dimension, where the doubling dimension is the least value α such that any ball in the network can be covered by at most 2 α balls of half radius. There are two variants of routing-scheme design: (i) labeled (name-dependent) routing, in which the designer is allowed to rename the nodes so that the names (labels) can contain additional routing information, for example, topological information; and (ii) name-independent routing, which works on top of the arbitrary original node names in the network, that is, the node names are independent of the routing scheme. In this article, given any constant ε ϵ (0, 1) and an n-node edge-weighted network of doubling dimension α ϵ O(loglog n), we present —a (1 + ε)-stretch labeled compact routing scheme with Γlog n-bit routing labels, O(log 2 n/loglog n)-bit packet headers, and ((1/ε) O(α) log 3 n)-bit routing information at each node; —a (9 + ε)-stretch name-independent compact routing scheme with O(log 2 n/loglog n)-bit packet headers, and ((1/ε) O(α) log 3 n)-bit routing information at each node. In addition, we prove a lower bound: any name-independent routing scheme with o(n (ε/60)2 ) bits of storage at each node has stretch no less than 9 - ε for any ε ϵ (0, 8). Therefore, our name-independent routing scheme achieves asymptotically optimal stretch with polylogarithmic storage at each node and packet headers. Note that both schemes are scale-free in the sense that their space requirements do not depend on the normalized diameter Δ of the network. Finally, we also present a simpler nonscale-free (9 + ε)-stretch name-independent compact routing scheme with improved space requirements if Δ is polynomial in n.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Machine-Learning-Based Monitoring of Laser Powder Bed Fusion

A two-step machine learning approach to monitoring Laser Powder Bed Fusion (LPBF) additive manufacturing is demonstrated that enables on-the-fly assessments of laser track welds. First, in situ video melt pool data acquired during LPBF is labeled according to the (1) average and (2) standard deviation of individual track width and also (3) whether or not the track is continuous, measured post-build through an ex situ height map analysis algorithm. This procedure generates three ground truth labeled datasets for supervised machine learning. Using a portion of the labeled 10-millisecond video clips, a single Convolutional Neural Network architecture is trained to generate three distinct networks. With the remaining in situ LPBF data, the trained neural networks are tested and evaluated and found to predict track width, standard deviation, and continuity without the need for ex situ measurements. This two-step approach should benefit any LPBF system – or any additive manufacturing technology – where height-map-derived properties can serve as useful labels for in situ sensor data.

42 ENGINEERING↗
DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗
DOE OSTI2018

Isotopically nonstationary metabolic flux analysis (INST-MFA): putting theory into practice

Typically, 13 C flux analysis relies on assumptions of both metabolic and isotopic steady state. If metabolism is steady but isotope labeling is not allowed to fully equilibrate, isotopically nonstationary metabolic flux analysis (INST-MFA) can be used to estimate fluxes. This requires solution of differential equations that describe the time-dependent labeling of network metabolites, while iteratively adjusting the flux and pool size parameters to match the transient labeling measurements. INST-MFA holds a number of unique advantages over approaches that rely solely upon steady-state isotope enrichments. First, INST-MFA can be applied to estimate fluxes in autotrophic systems, which consume only single-carbon substrates. Second, INST-MFA is ideally suited to systems that label slowly due to the presence of large intermediate pools or pathway bottlenecks. Finally, INST-MFA provides increased measurement sensitivity to estimate reversible exchange fluxes and metabolite pool sizes, which represents a potential framework for integrating metabolomic analysis with 13 C flux analysis. This review highlights the unique capabilities of INST-MFA, describes newly available software tools that automate INST-MFA calculations, presents several practical examples of recent INST-MFA applications, and discusses the technical challenges that lie ahead.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Active Learning for Anomaly Detection in Environmental data

Due to the growing amount of data from in-situ sensors in environmental monitoring, it becomes necessary to automatically detect anomalous data points. Nowadays, this is mainly performed using supervised machine learning models, which need a fully labelled data set for their training process. However, the process of labelling data is typically cumbersome and, as a result, a hindrance to the adoption of machine learning methods for automated anomaly detection. In this work, we propose to address this challenge by means of active learning. This method consists of querying the domain expert for the labels of only a selected subset of the full data set. We show that this reduces the time and costs associated to labelling while delivering the same or similar anomaly detection performances. Finally, we also show that machine learning models providing a nonlinear classification boundary are to be recommended for anomaly detection in complex environmental data sets.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Sim-to-real supervised domain adaptation for radioisotope identification

Machine learning has the potential to improve the speed and reliability of radioisotope identification using gamma spectroscopy. However, meticulously labeling an experimental dataset for training is often prohibitively expensive, while training models purely on synthetic data is risky due to the domain gap between simulated and experimental measurements. In this research, we demonstrate that supervised domain adaptation can substantially improve the performance of radioisotope identification models by transferring knowledge between synthetic and experimental data domains. We consider two domain adaptation scenarios: (1) a simulation-to-simulation adaptation, where we perform multi-label proportion estimation using simulated high-purity germanium detectors, and (2) a simulation-to-experimental adaptation, where we perform multi-class, single-label classification using measured spectra from handheld lanthanum bromide (LaBr) and sodium iodide (NaI) detectors. We begin by pretraining a spectral classifier on synthetic data using a custom transformer-based neural network. After subsequent fine-tuning on just 64 labeled experimental spectra, we achieve a test accuracy of 96% in the sim-to-real scenario with a LaBr detector, far surpassing a synthetic-only baseline model (75%) and a model trained from scratch (80%) on the same 64 spectra. Furthermore, we demonstrate that domain-adapted models learn more human-interpretable features than experiment-only baseline models. Overall, our results highlight the potential for supervised domain adaptation techniques to bridge the sim-to-real gap in radioisotope identification, enabling the development of accurate and explainable classifiers even in real-world scenarios where access to experimental data is limited.

Lalor, Peter W.↗
DOE OSTI2025

Lipids represent a dynamic, yet stable pool of microbially-derived soil carbon

There is an emerging consensus that microorganisms are a primary source of persistent, slow-cycling soil organic matter (SOM), however which microbial residues contribute to SOM and what controls their accumulation remains unresolved. Lipids are a commonly overlooked biomolecular pool that could contribute significantly to stable SOM. While current estimates for phospholipid degradation in soils are rapid, lipids are structurally heterogeneous molecules that could turnover at different rates. Through a year-long soil incubation and compound-specific, stable isotope probing (SIP)-lipidomics, we were able to rigorously track the persistence of lipid compounds in silty and sandy switchgrass bioenergy crop soils. Here, we assessed the influence of lipid structure on soil lipid accrual and degradation as moderated by soil texture, since mineral association is presumed to be the primary mechanism for lipid persistence. After rapid incorporation of 13 C glucose into microbial biomass, we found 13 C label was retained broadly across chemically diverse lipid classes, even after one year. 13 C-labeled lipid profiles varied significantly with soil texture; however, we found no difference between sandy and silty soils in lipid retention, suggesting soil texture may only play a minor role in modulating lipid persistence. Only two lipid subclasses were found to be persistent (i.e., retention of 13 C label without significant degradation or production): phosphatidylinositol lipids and hydroxyceramide lipids, both of which are negatively charged, possibly facilitating stabilization by mineral complexation. However, several other subclasses displayed substantial ongoing production. In particular, the accumulation of triacylglycerol lipids across soil textures suggests that storage lipids may be an important component of SOC, highlighting a potential target for management strategies to promote C retention by lipid accrual. Overall, the retention for over a year of 13 C label in microbial intact lipid biomarkers reveals the importance of efficient biomass production and turnover when considering microbial C contributions to SOM.

compound-specific↗
DOE OSTI2026

NMR of Fully and Partially 13 C-Enriched Biomass Enhances Pendent Group Structural Characterization

Traditional solution-state NMR experiments may either fail or yield unsatisfactory results when employing fully- 13 C-labeled biomass due to complications arising from 13 C– 13 C coupling. Constant-time analogs of HSQC experiments mitigate such issues and deliver enhanced sensitivity. A rarely reported CT-HSQC-TOCSY experiment allows the proton coupling network to deliver much of the same value as the parent experiment on unlabeled or 10–15%- 13 C-labeled biomass polymers but with enhanced sensitivity. In the absence of a viable HMBC analog for long-range correlations, a relayed C–C experiment, i.e., via directly bonded 13 C-labeled networks, enables the reliable assignment of coupled carbons, with the added advantage of correlating the more elusive quaternaries. A C–C-FLOPSY experiment takes advantage of fully- 13 C-labeled materials for mapping extensive carbon networks in the complex polymer mixtures inherent in biomass. Various pendent groups (tricin units, cis- and trans-p-coumarates, and p -hydroxybenzoates) that adorn lignins, and the cis- and trans-ferulates on arabinoxylan polysaccharides, are exquisitely revealed in spectra from isolated lignins or whole-cell-wall materials from maize, sorghum, and poplar.

biopolymers↗