Search NASA⌕ Search

SEARCH · Search NASA

Results for “binary classification”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

NukeLM: Pre-Trained and Fine-Tuned Language Models for the Nuclear and Energy Domains

Natural language processing (NLP) tasks (text classification, named entity recognition, etc.) have seen amazing improvements over the last few years. This is due to models such as BERT that achieve deep knowledge transfer by using a large pre-trained model, then fine-tuning the model on specific tasks. The BERT architecture has shown even better performance on domain-specific tasks when the model is pre-trained using domain-relevant texts. Here, inspired by these recent advancements, we have developed NukeLM, a nuclear-domain BERT model pre-trained on 1.5 million abstracts from the DOE Office of Scientific and Technical Information (OSTI) database. This NukeLM model is then fine-tuned for the classification of research articles into either binary classes (related to the nuclear fuel cycle (NFC) or not) or multiple categories related to the subject of the article. We show that continued pre-training of a BERT-style architecture prior to fine-tuning results in greater performance in both article classification tasks. This information is critical for properly triaging manuscripts, a necessary task for better understanding citation networks that publish in the nuclear space and uncovering new areas of research in the nuclear (or nuclear relevant) domain.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2017

The X-Ray Binary Population of the Nearby Dwarf Starburst Galaxy IC 10: Variable and Transient X-Ray Sources

We have monitored the Cassiopeia dwarf galaxy (IC 10) in a series of 10 Chandra ACIS-S observations to capture its variable and transient X-ray source population, which is expected to be dominated by High Mass X-ray Binaries (HMXBs). We present a sample of 21 X-ray sources that are variable between observations at the 3 σ level, from a catalog of 110 unique point sources. We find four transients (flux variability ratio greater than 10) and a further eight objects with ratios >5. The observations span the years 2003–2010 and reach a limiting luminosity of >10{sup 35} erg s{sup −1}, providing sensitivity to X-ray binaries in IC 10 as well as flare stars in the foreground Milky Way. The nature of the variable sources is investigated from light curves, X-ray spectra, energy quantiles, and optical counterparts. The purpose of this study is to discover the composition of the X-ray binary population in a young starburst environment. IC 10 provides a sharp contrast in stellar population age (<10 My) when compared to the Magellanic Clouds (40–200 My) where most of the known HMXBs reside. We find 10 strong HMXB candidates, 2 probable background Active Galactic Nuclei, 4 foreground flare-stars or active binaries, and 5 not yet classifiable sources. Complete classification of the sample requires optical spectroscopy for radial velocity analysis and deeper X-ray observations to obtain higher S/N spectra and search for pulsations. A catalog and supporting data set are provided.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Data supporting manuscript from L. Sheneman, G. Stephanopoulos, A.E. Vasdekis titled "Deep learning classification of lipid droplets in quantitative phase images" as currently under review at PLOS ONE. This includes: 1) raw and binary labeled Quantitative Phase Images (QPI) of Y. lipolytica cells used in the analyses described within the manuscript. 2) various derived data including classifier scores, etc.

Data supporting manuscript from L. Sheneman, G. Stephanopoulos, A.E. Vasdekis titled "Deep learning classification of lipid droplets in quantitative phase images" as currently under review at PLOS ONE. This includes: 1) raw and binary labeled Quantitative Phase Images (QPI) of Y. lipolytica cells used in the analyses described within the manuscript. 2) various derived data including classifier scores, etc.

ANN↗
DOE OSTI2013

Identification of five interacting binaries in the galactic bulge survey

We present optical light curves, spectroscopy, and classification of five X-ray sources in the Chandra Galactic Bulge Survey (CXOGBS J174009.1–284725 (CX5), CXOGBS J173935.7–272935 (CX18), CXOGBS J173946.9–271809 (CX28), CXOGBS J173729.1–292804 (CX37), CXOGBS J174607.6–261547 (CX561)). These objects were selected based on bright optical counterparts which were quickly found to have emission lines in their optical spectra. This paper presents an illustration of GBS optical follow-up, targeting emission line objects. Of these five objects, four exhibit photometric variability in the Sloan r' band. CX5 shows a tentative period of 2.1 hr and is clearly an intermediate polar (IP). CX28 and CX37 both exhibit flickering with no clear period. Both are also suggested to be IPs. CX18 was observed to undergo two dwarf nova outbursts. Finally, CX561 shows no detectable variability, although its characteristics would be consistent with either a quiescent low-mass X-ray binary or cataclysmic variable.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Adversarial Binaries: AI-guided Instrumentation Methods for Malware Detection Evasion

Adversarial binaries are executable files that have been altered without loss of function by an AI agent in order to deceive malware detection systems. Progress in this emergent vein of research has been constrained by the complex and rigid structure of executable files. Although prior work has demonstrated that these binaries deceive a variety of malware classification models which rely on disparate feature sets, a consensus as to the best approach has not been reached, either in terms of the optimization algorithms or the instrumentation methods. Furthermore, although inconsistencies in the data sets, target classifiers, and functionality verification methods make head-to-head comparisons difficult, here we extract lessons learned and make recommendations for future research.

malware obfuscation↗
DOE OSTI2024

CMOS-Based Single-Cycle in-Memory XOR/XNOR

Big data applications are on the rise, and so is the number of data centers. The ever-increasing massive data pool needs to be periodically backed up in a secure environment. Moreover, a massive amount of securely backed-up data is required for training binary convolutional neural networks for image classification. XOR and XNOR operations are essential for large-scale data copy verification, encryption, and classification algorithms. The disproportionate speed of existing compute and memory units makes the von Neumann architecture inefficient to perform these Boolean operations. Compute-in-memory (CiM) has proved to be an optimum approach for such bulk computations. The existing CiM-based XOR/XNOR techniques either require multiple cycles for computing or add to the complexity of the fabrication process. Here, we propose a CMOS-based hardware topology for single-cycle in-memory XOR/XNOR operations. Our design provides at least 2× improvement in the latency compared with other existing CMOS-compatible solutions. We verify the proposed system through circuit/system-level simulations and evaluate its robustness using a 5000-point Monte Carlo variation analysis. This all-CMOS design paves the way for practical implementation of CiM XOR/XNOR at scaled technology nodes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Leveraging large language models to address data scarcity in machine learning for graphene synthesis

Machine learning in experimental materials science faces significant challenges due to the scarcity of data, which are costly and time-consuming to generate, particularly when relying on in-house experiments. Literature data mining offers a potential solution but introduces issues like mixed data quality, inconsistent formats, and non-uniform reporting of synthesis parameters, resulting in partially missing and heterogeneous features across the dataset. Here, we propose data imputation and feature engineering methods that employ pre-trained large language models (LLMs) to enhance machine learning performance on scarce, heterogeneous datasets, demonstrated on graphene CVD synthesis data and the ML-HydPARK hydrogen storage dataset. GPT models perform data imputation via tailored prompting and semantic normalization of inconsistently reported features through embeddings, for example, to harmonize the complex nomenclature of CVD substrates. Beyond yielding more diverse and richer feature representations than traditional methods such as K-nearest neighbors (KNN) and Multivariate Imputation by Chained Equations (MICE), LLM-based data imputation is evaluated against dataset characteristics and prompting strategies. We vary the level of autonomy granted to the LLM, from generic prompting that leverages pre-trained knowledge for autonomous data generation to data-informed prompting that constrains outputs using target-specific information, and demonstrate which level of autonomy yields superior imputation performance across datasets and feature types. The proposed data engineering methods markedly improve downstream performance; for example, in graphene layer number classification using a support vector machine (SVM), binary accuracy increases from 39% to 65% and ternary accuracy from 52% to 72%. Fine-tuning experiments on both datasets show that combining our proposed LLM-based data imputation and feature encoding methods with numerical machine learning predictors outperforms standalone fine-tuned LLM predictors in data-scarce settings. The proposed strategies emphasize data enhancement techniques rather than refining learning architectures or regularizing loss functions, offering a broadly applicable framework for improving machine learning performance on scarce, inhomogeneous datasets.

Chemical vapor deposition↗
DOE OSTI2023

Direct analysis of the broad-line SN 2019ein: connection with the core-normal SN 2011fe

Type Ia supernovae (SNe Ia) are important cosmological probes and contributors to galactic nucleosynthesis, particularly of the iron group elements. To improve both their reliability as cosmological probes and to understand galactic chemical evolution, it is vital to understand the binary progenitor system and explosion mechanism. The classification of SNe Ia into Branch groups has led to some understanding of the similarities and differences among the varieties of observed SNe Ia. Branch groups are defined by the pseudo-equivalent widths of the two prominent Si ii lines, leading to four distinct groups: core-normal (CN), shallow-silicon, cool, and broad-line (BL). However, partly due to small sample size, little work has been done on the BL group. We perform direct spectral analysis on the pre-maximum spectra of the BL SN 2019ein, comparing and contrasting to the CN SN 2011fe. Both SN 2019ein and SN 2011fe were first observed spectroscopically within 2 d of discovery, allowing us to follow the spectroscopic evolution of both SNe in detail. We find that the optical depths of the primary features of both the CN and BL SNe are very similar, except that there is a Doppler shift between them. We further examine the BL group and show that for nine objects with pre-maximum spectra in the range (–6)–(–2) d with respect to B-maximum all the emission peaks of the Si ii λ6355 line of BL are blueshifted pre-maximum, suggesting a possible classification criterion.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

THE GALACTIC O-STAR SPECTROSCOPIC SURVEY (GOSSS). III. 142 ADDITIONAL O-TYPE SYSTEMS

This is the third installment of the Galactic O-Star Spectroscopic Survey (GOSSS), a massive spectroscopic survey of Galactic O stars, based on new homogeneous, high signal-to-noise ratio, R ∼ 2500 digital observations selected from the Galactic O-Star Catalog. In this paper, we present 142 additional stellar systems with O stars from both hemispheres, bringing the total of O-type systems published within the project to 590. Among the new objects, there are 20 new O stars. We also identify 11 new double-lined spectroscopic binaries, 6 of which are of O+O type and 5 of O+B type, and an additional new tripled-lined spectroscopic binary of O+O+B type. We also revise some of the previous GOSSS classifications, present some egregious examples of stars erroneously classified as O-type in the past, introduce the use of luminosity class IV at spectral types O4-O5.5, and adapt the classification scheme to the work of Arias et al.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

GRB 081024B and GRB 140402A: Two Additional Short GRBs from Binary Neutron Star Mergers

Theoretical and observational evidences for a two-fold classification of short bursts have been recently obtained: (1) short gamma-ray flashes (S-GRFs), with isotropic energy E{sub iso}<10{sup 52} erg and no black hole (BH) formation, and (2) authentic short gamma-ray bursts (S-GRBs), with isotropic energy E{sub iso}>10{sup 52} erg showing evidence of BH formation in the binary neutron star merging process. The signature for BH formation is the onset of high-energy (0.1–100 GeV) emission, coeval to the prompt emission, in all S-GRBs. No GeV emission is expected nor observed in S-GRFs. In this paper, we present two S-GRBs, GRB 081024B and GRB 140402A, in addition to the already identified S-GRBs, GRB 090227B, GRB 090510, and GRB 140619B. We also return to the absence of GeV emission in the S-GRB 090227B, at an angle of 71{sup ∘} from the Fermi-LAT boresight. All of the correctly identified S-GRBs correlate with high-energy emission, implying no significant presence of beaming in GeV emission. The existence of a common power-law behavior in the GeV luminosities, following the BH formation, when measured in the source rest frame, points to a commonality in the mass and spin of the newly formed BHs in all S-GRBs.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Topological approach to electron correlations at fractional quantum Hall effect

Highlights: • Braids in 2D electron systems in magnetic field acquire a cyclotron metrics. • Commensurability of 2D braids with Wigner crystal of electrons leads to FQHE. • Homotopy invariants define the hierarchy of FQHE universal in all 2D Hall systems. • Composite fermions illustrate multiloop braids in the simplest homotopy case. • Correlations in FQHE reveal long-range quantum entanglement of all electrons. The classification of homotopy invariants in interacting multi-electron 2D systems at quantizing magnetic fields is presented, explaining the topologically protected correlations occurring at integer and fractional quantum Hall effects. The long-range quantum entanglement is essential for homotopy correlated phases in contrast to the binary entanglement for conventional phases with local order parameters. The classification of homotopy long-range correlated phases induced by the Coulomb interaction of electrons has been derived in terms of homotopy invariants, which are universal and robust against local disorder and single-particle crystal field, as illustrated by experimental observations in various materials with different microscopic structure, like GaAs 2DES, graphene monolayer and bilayer and in Chern topological insulators. The homotopy phases are demonstrated to be topologically protected and immune to single-particle perturbations, temperature chaos and variation of the electron interaction strength. The nonzero repulsive interaction between electrons is shown, however, to be essential for the definition of the homotopy invariants, which disappear in gaseous systems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Vulcan-Forge: Architecture and Design of a Multi-Modal Forensic Analysis Plugin for CALDERA

Forge and VULCAN together describe an open-architecture cybersecurity analysis ecosystem that unifies forensic artifact processing, detection engineering, and vulnerability intelligence within integrated platforms. Forge operates as a plugin for MITRE CALDERA, ingesting diverse evidence formats—including EVTX, PCAP/PCAPNG, CSV, JSON, YAML, XML, binaries, and archives—to construct a unified artifact graph enriched with severity scoring, TLP classification, and audit trails. It provides subsystems for artifact parsing, streaming structured-data visualization, NetworkMiner-based packet inspection, PE/.NET binary analysis, and LLM-assisted triage and rule generation, with outputs validated against CCCS-YARA and pySigma schemas. VULCAN complements this by serving as a cybersecurity analyst platform that integrates a Neo4j knowledge graph, Qdrant vector retrieval, SSVC-based triage, and a local LLM to deliver CVE intelligence and forensic analysis through a multi-source ingest pipeline drawing from NVD, CISA KEV, EPSS, MITRE ATT&CK, and CAPEC. Together, they bridge structured threat intelligence with automated forensic analysis and detection workflows.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

On the classification of GRBs and their occurrence rates

There is mounting evidence for the binary nature of the progenitors of gamma-ray bursts (GRBs). For a long GRB, the induced gravitational collapse paradigm proposes as progenitor, or “in-state,” a tight binary system composed of a carbon–oxygen core (CO{sub core}) undergoing a supernova explosion that triggers hypercritical accretion onto a neutron star (NS) companion. For a short GRB (S-GRB), an NS–NS merger is traditionally adopted as the progenitor. We divide long and S-GRBs into two subclasses, depending on whether or not a black hole (BH) is formed in the merger or in the hypercritical accretion process exceeding the NS critical mass. For long bursts, when no BH is formed, we have the subclass of X-ray flashes (XRFs), with isotropic energy E{sub iso}≲10{sup 52} erg and rest-frame spectral peak energy E{sub p,i}≲200 keV. When a BH is formed, we have the subclass of binary-driven hypernovae (BdHNe), with E{sub iso}≳10{sup 52} erg and E{sub p,i}≳200 keV. In analogy, short bursts are similarly divided into two subclasses. When no BH is formed, short gamma-ray flashes (S-GRFs) occur, with E{sub iso}≲10{sup 52} erg and E{sub p,i}≲2 MeV. When a BH is formed, the authentic S-GRBs occur, with E{sub iso}≳10{sup 52} erg and E{sub p,i}≳2 MeV. We give examples and observational signatures of these four subclasses and their rate of occurrence. From their respective rates it is possible that “in-states” of S-GRFs and S-GRBs originate from the “out-states” of XRFs. We indicate two additional progenitor systems: white dwarf–NS and BH–NS. These systems have hybrid features between long and short bursts. In the case of S-GRBs and BdHNe evidence is given of the coincidence of the onset of the high-energy GeV emission with the birth of a Kerr BH.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

PLASMA-WAVE GENERATION IN A DYNAMIC SPACETIME

We propose a new electromagnetic (EM)-emission mechanism in magnetized, force-free plasma, which is driven by the evolution of the underlying dynamic spacetime. In particular, the emission power and angular distribution of the emitted fast-magnetosonic and Alfvén waves are separately determined. Previous numerical simulations of binary black hole mergers occurring within magnetized plasma have recorded copious amounts of EM radiation that, in addition to collimated jets, include an unexplained, isotropic component that becomes dominant close to the merger. This raises the possibility of multimessenger gravitational-wave and EM observations on binary black hole systems. The mechanism proposed here provides a candidate analytical characterization of the numerical results, and when combined with previously understood mechanisms such as the Blandford–Znajek process and kinetic-motion-driven radiation, it allows us to construct a classification of different EM radiation components seen in the inspiral stage of compact-binary coalescences.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Training a Quantum Annealing Based Restricted Boltzmann Machine on Cybersecurity Data

A restricted Boltzmann machine (RBM) is a generative model that could be used in effectively balancing a cybersecurity dataset because the synthetic data a RBM generates follows the probability distribution of the training data. RBM training can be performed using contrastive divergence (CD) and quantum annealing (QA). QA-based RBM training is fundamentally different from CD and requires samples from a quantum computer. We present a real-world application that uses a quantum computer. Specifically, we train a RBM using QA for cybersecurity applications. The D-Wave 2000Q has been used to implement QA. RBMs are trained on the ISCX data, which is a benchmark dataset for cybersecurity. For comparison, RBMs are also trained using CD. CD is a commonly used method for RBM training. Our analysis of the ISCX data shows that the dataset is imbalanced. We present two different schemes to balance the training dataset before feeding it to a classifier. The first scheme is based on the undersampling of benign instances. The imbalanced training dataset is divided into five sub-datasets that are trained separately. A majority voting is then performed to get the result. Our results show the majority vote increases the classification accuracy up from 90.24% to 95.68%, in the case of CD. For the case of QA, the classification accuracy increases from 74.14% to 80.04%. In the second scheme, a RBM is used to generate synthetic data to balance the training dataset. We show that both QA and CD-trained RBM can be used to generate useful synthetic data. Balanced training data is used to evaluate several classifiers. Among the classifiers investigated, K-Nearest Neighbor (KNN) and Neural Network (NN) perform better than other classifiers. They both show an accuracy of 93%. Our results show a proof-of-concept that a QA-based RBM can be trained on a 64-bit binary dataset. The illustrative example suggests the possibility to migrate many practical classification problems to QA-based techniques. Further, we show that synthetic data generated from a RBM can be used to balance the original dataset.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

The Chandra local volume survey. I. The X-ray point source populations of NGC 55, NGC 2403, and NGC 4214

We present comprehensive X-ray point source catalogs of NGC 55, NGC 2403, and NGC 4214 as part of the Chandra Local Volume Survey. The combined archival observations have effective exposure times of 56.5 ks, 190 ks, and 79 ks for NGC 55, NGC 2403, and NGC 4214, respectively. When combined with our published catalogs for NGC 300 and NGC 404, our survey contains 629 X-ray sources total down to a limiting unabsorbed luminosity of ∼5×10{sup 35} erg s{sup −1} in the 0.35–8 keV band in each of the five galaxies. We present X-ray hardness ratios, spectral analysis, radial source distributions, and an analysis of the temporal variability for the X-ray sources detected at high significance. To constrain the nature of each X-ray source, we carried out cross-correlations with multi-wavelength data sets. We searched overlapping Hubble Space Telescope observations for optical counterparts to our X-ray detections to provide preliminary classifications for each X-ray source as a likely X-ray binary, background active galactic nucleus, supernova remnant, or foreground star.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Orbital Dynamics of Candidate Transitional Millisecond Pulsar 3FGL J1544.6-1125: An Unusually Face-on System

We present the orbital solution for the donor star of the candidate transitional millisecond pulsar (tMSP) 3FGL J1544.6-1125, currently observed as an accreting low-mass X-ray binary. The orbital period is 0.2415361(36) days, entirely consistent with the spectral classification of the donor star as a mid-to-late K dwarf. The semi-amplitude of the radial velocity curve is exceptionally low at K{sub 2}=39.3±1.5 km s{sup −1}, implying a remarkably face-on inclination in the range 5°–8°, depending on the neutron star and donor masses. After determining the veiling of the secondary, we derive a distance to the binary of 3.8 ± 0.7 kpc, yielding a 0.3–10 keV X-ray luminosity of 6.1±1.9×10{sup 33} erg s{sup −1}, similar to confirmed tMSPs. As face-on binaries rarely occur by chance, we discuss the possibility that Fermi-selected samples of transitional millisecond pulsars in the sub-luminous disk state are affected by beaming. By phasing emission line strength on the spectroscopic ephemeris, we find coherent variations, and argue that these variations are most consistent with emission from an asymmetric shock originating near the inner disk.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2012

THE CHANDRA LOCAL VOLUME SURVEY: THE X-RAY POINT-SOURCE CATALOG OF NGC 300

We present the source catalog of a new Chandra ACIS-I observation of NGC 300 obtained as part of the Chandra Local Volume Survey. Our 63 ks exposure covers {approx}88% of the D{sub 25} isophote (R Almost-Equal-To 6.3 kpc) and yields a catalog of 95 X-ray point sources detected at high significance to a limiting unabsorbed 0.35-8 keV luminosity of {approx}10{sup 36} erg s{sup -1}. Sources were cross-correlated with a previous XMM-Newton catalog, and we find 75 'X-ray transient candidate' sources that were detected by one observatory, but not the other. We derive an X-ray scale length of 1.7 {+-} 0.2 kpc and a recent star formation rate of 0.12 M{sub Sun} yr{sup -1} in excellent agreement with optical observations. Deep, multi-color imaging from the Hubble Space Telescope, covering {approx}32% of our Chandra field, was used to search for optical counterparts to the X-ray sources, and we have developed a new source classification scheme to determine which sources are likely X-ray binaries, supernova remnants, and background active galactic nucleus candidates. Finally, we present the X-ray luminosity functions (XLFs) at different X-ray energies, and we find the total NGC 300 X-ray point-source population to be consistent with other late-type galaxies hosting young stellar populations ({approx}< 50 Myr). We find that XLF of sources associated with older stellar populations has a steeper slope than the XLF of X-ray sources coinciding with young stellar populations, consistent with theoretical predictions.

79 ASTRONOMY AND ASTROPHYSICS↗