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.

36 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

ICAT: The Interactive Corpus Analysis Tool

The Interactive Corpus Analysis Tool (ICAT) is a Python library for creating dashboards to explore textual datasets and build simple binary classification models to help filter through them and focus on entries of interest. This tool uses a form of interactive machine learning (IML), a paradigm of “machine teaching” (Simard et al., 2017) that sits at the intersection of the fields of human computer interaction (HCI), visual analytics, and machine learning. The intent of ICAT is to allow subject matter experts (SME) with limited to no experience in machine learning to benefit from an iterative human-in-the-loop (HITL) approach to building their own model without needing to understand the details of the underlying algorithm. This interactivity is achieved by allowing the user to create features, label data points, and visually manipulate a representation of the features to manually cluster and investigate data, while a model is trained on the fly based on these actions. ICAT is built on top of the Panel (Holoviz, 2018) library, using a combination of Vega, a custom IPyWidget using D3, and ipyvuetify, and is intended to be used inside of a Jupyter environment.

Martindale, Nathan [Oak Ridge National Laboratory ↗
DOE OSTI2025

Performance Evaluation of Vertical Federated Machine Learning Against Adversarial Threats on Wide-Area Control System: Preprint

Federated machine learning (FL) is gaining significant popularity to develop cybersecurity solutions in power grids because of its advanced capability to support decentralized data handing at local devices, its privacy preservation, and its low-bandwidth requirement. However, the evolving adversarial machine learning (AML) threats raise significant concerns for the cybersecurity of FL architectures. The FL-based split neural network (SplitNN) achieves high performance through the decentralized training of local neural network models while preserving data privacy across multiple entities. In this paper, we propose a methodology for evaluating the performance of a vertical FLbased anomaly detector against different types of AML attacks, including denial-of-service attacks, adversarial data injection attacks, and replay attacks on the trained local models deployed in the grid network. For a case study, we consider the modified IEEE 13-bus system, and we develop SplitNN-based binary and multiclass classification models to detect, locate, and identify different types of data integrity attacks on the volt-watt control with two pooling layers: maximum pooling and AvgPool. Our experimental results, computed through performance metrics, reveal that the severity of these AML attacks varies with the integrated pooling mechanism, the type of classification model, and the nature of the cyberattack. Further, the AML attacks negatively impacted the prediction time per sample for the pretrained SplitNN during the online testing.

adversarial threats↗
DOE OSTI2026

A unified large language model–based framework for heterogeneous PV image diagnosis

With advances in imaging technologies, modern photovoltaic (PV) systems generate large volumes of heterogeneous image data, including visible, electroluminescence (EL), and infrared (IR) images. Existing PV image analysis models, particularly deep learning approaches, are typically task-specific and lack cross-modality generalization. To address this limitation, this paper proposes an open-source large language model (LLM)–based unified framework for heterogeneous PV image diagnostics. Through task-aware diagnostic prompting, the framework enables analysis of visible, EL, and IR images within a single pipeline, supporting both zero-shot and few-shot inference and binary and multiclass classification. It is compatible with state-of-the-art multimodal LLMs, including ChatGPT, Gemini, Claude, Qwen, and CLIP. The framework is evaluated on PV module condition classification (clean, soiling, snow, hail, and bird droppings) using visible images, cell crack detection using EL images, and hotspot detection using IR images. GPT-5.1 in few-shot mode achieves the best performance, with classification accuracy exceeding 97.3%. Open-source models such as Qwen and CLIP also deliver competitive results on visible images (around 90% accuracy), though their performance is more limited on EL and IR modalities. On the full ELPV dataset, the framework achieves 83.5% zero-shot accuracy, within 2.8% of the supervised CNN baseline, confirming scalability to larger benchmarks. Practical aspects such as reproducibility, response latency, and confidence estimation are systematically analyzed. The framework operates across PV image modalities without modality- or task-specific training, making it well suited as a rapid pre-screening tool to support downstream detailed diagnostics. A benchmark dataset of diverse labeled PV images is also released.

Li, Baojie↗
DOE OSTI2024

The DESI Early Data Release white dwarf catalogue

The Early Data Release (EDR) of the Dark Energy Spectroscopic Instrument (DESI) comprises spectroscopy obtained from 2020 December 14 to 2021 June 10. White dwarfs were targeted by DESI both as calibration sources and as science targets and were selected based on Gaia photometry and astrometry. Here, we present the DESI EDR white dwarf catalogue, which includes 2706 spectroscopically confirmed white dwarfs of which approximately 60 per cent have been spectroscopically observed for the first time, as well as 66 white dwarf binary systems. We provide spectral classifications for all white dwarfs, and discuss their distribution within the Gaia Hertzsprung–Russell diagram. We provide atmospheric parameters derived from spectroscopic and photometric fits for white dwarfs with pure hydrogen or helium photospheres, a mixture of those two, and white dwarfs displaying carbon features in their spectra. We also discuss the less abundant systems in the sample, such as those with magnetic fields, and cataclysmic variables. The DESI EDR white dwarf sample is significantly less biased than the sample observed by the Sloan Digital Sky Survey, which is skewed to bluer and therefore hotter white dwarfs, making DESI more complete and suitable for performing statistical studies of white dwarfs.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Chandra Discovery of a Candidate Hyperluminous X-Ray Source in MCG+11-11-032

We present a multiwavelength analysis of MCG+11-11-032, a nearby active galactic nucleus (AGN), with a unique classification as being both a binary and a dual AGN candidate. With new Chandra observations, we aim to resolve any dual AGN system via imaging data and search for signs of a binary AGN via analysis of the X-ray spectrum. Analyzing the Chandra spectrum, we find no evidence of the previously suggested double-peaked Fe Kα lines; the spectrum is instead best fit by an absorbed power law with a single Fe Kα line, as well as an additional line centered at ≈7.5 keV. The Chandra observation reveals faint, soft, and extended X-ray emission, possibly linked to low-level nuclear outflows. Further analysis shows evidence for a compact hard source—MCG+11-11-032 X2—located 3.″3 from the primary AGN. Modeling MCG+11-11-032 X2 as a compact source, we find that it is relatively luminous (L2–10 keV=1.5−0.5+0.9×1041erg s$^{−1}$), and the location is coincident with a compact and off-nuclear source resolved in Hubble Space Telescope infrared (F105W) and optical (F621M, F547M) bands. Pairing our X-ray results with a 144 MHz radio detection at the host galaxy location, we observe X-ray and radio properties similar to those of ESO 243-49 HLX-1, suggesting that MCG+11-11-032 X2 may be a hyperluminous X-ray source. This detection with Chandra highlights the importance of a high-resolution X-ray imager as well as how previous binary AGN candidates detected with large-aperture instruments can benefit from high-resolution follow-up. Future spatially resolved optical spectra, and deeper X-ray observations, can better constrain the origin of MCG+11-11-032 X2.

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 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 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 OSTI2025

Scaling laws in jet classification

We demonstrate the emergence of scaling laws in the benchmark top versus QCD jet classification problem in collider physics. Six distinct physically-motivated classifiers exhibit power-law scaling of the binary cross-entropy test loss as a function of training set size, with distinct power law indices. This result highlights the importance of comparing classifiers as a function of dataset size rather than for a fixed training set, as the optimal classifier may change considerably as the dataset is scaled up. We speculate on the interpretation of our results in terms of previous models of scaling laws observed in natural language and image datasets.

Batson, Joshua↗
DOE OSTI2026

VHClass

The code is used to predict the taxonomic source of an antibody heavy chain sequence. The code assigns a binary label to the input set of sequences - camelid or human. This prediction is generated using a random-forest based classification algorithm which is the backbone of the code. A complementary code splits the antibody sequence into antibody features - framework regions and CDR regions.

Davis, Anastasiia↗
DOE OSTI2025

A knowledge-informed large language model framework for U.S. nuclear power plant shutdown initiating event classification for probabilistic risk assessment

Identifying and classifying shutdown initiating events (SDIEs) is critical for developing shutdown probabilistic risk assessment for nuclear power plants. Existing computational approaches cannot achieve satisfactory performance due to the challenges of unavailable large, labeled datasets, imbalanced event types, and label noise. To address these challenges, we propose a hybrid pipeline that integrates a knowledge-informed machine learning model to prescreen non-SDIEs and a large language model (LLM) to classify SDIEs into four types. In the prescreening stage, we proposed a set of 44 SDIE text patterns that consist of the most salient keywords and phrases from six SDIE types. Text vectorization based on the SDIE patterns generates feature vectors that are highly separable by using a simple binary classifier. The second stage builds Bidirectional Encoder Representations from Transformers (BERT)-based LLM, which learns generic English language representations from self-supervised pretraining on a large dataset and adapts to SDIE classification by fine-tuning it on an SDIE dataset. The proposed approaches are evaluated on a dataset with 10,928 events using precision, recall ratio, F 1 score, and average accuracy. In conclusion, the results demonstrate that the prescreening stage can exclude more than 97% non-SDIEs, and the LLM achieves an average accuracy of 95.1% for SDIE classification.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2024

Probing the evolution of fault properties during the seismic cycle with deep learning

We use seismic waves that pass through the hypocentral region of the 2016 M6.5 Norcia earthquake together with Deep Learning (DL) to distinguish between foreshocks, aftershocks and time-to-failure (TTF). Binary and N-class models defined by TTF correctly identify seismograms in test with > 90% accuracy. We use raw seismic records as input to a 7 layer CNN model to perform the classification. Here we show that DL models successfully distinguish seismic waves pre/post mainshock in accord with lab and theoretical expectations of progressive changes in crack density prior to abrupt change at failure and gradual postseismic recovery. Performance is lower for band-pass filtered seismograms (below 10 Hz) suggesting that DL models learn from the evolution of subtle changes in elastic wave attenuation. Tests to verify that our results indeed provide a proxy for fault properties included DL models trained with the wrong mainshock time and those using seismic waves far from the Norcia mainshock; both show degraded performance. Our results demonstrate that DL models have the potential to track the evolution of fault zone properties during the seismic cycle. If this result is generalizable it could improve earthquake early warning and seismic hazard analysis.

58 GEOSCIENCES↗
DOE OSTI2026

Scalability Analysis of Quantum Models for Stress and Emotion Detection

Stress and emotion detection from high-dimensional physiological signals is a challenging task, particularly when aiming for accurate classification across diverse behavioral states. Quantum machine learning (QML) is promising for modeling such high-dimensional data, but scalability is limited by qubit resources and the exponential cost of classical statevector simulation. This work studies the scalability of quantum support vector machines (QSVMs) for binary stress detection and three-class emotion recognition (Negative/Neutral/Positive) under varying qubit counts and angle-encoding strategies. We also present a comparison study with one-feature-per-qubit (1:1) and two-features-per-qubit (2:1) mappings. Experiments are executed on HPC infrastructure using NVIDIA CUDA-Q to evaluate performance, variance, and class-dependent separability at higher-qubit setups. Results show that larger Hilbert spaces can improve peak accuracy but may increase instability. At the same time, dense 2:1 encoding yields more consistent stress detection performance. For emotion recognition, scaling improves discrimination for classes like Negative and Positive more than Neutral. We find that effective QML scaling is task-dependent and benefits more from encoding design than simply increasing qubit count.

Onim, Md. Saif Hassan [University of Tennessee, Kn↗
DOE OSTI2026

Two-Stage Wildlife Event Classification for Edge Deployment

Camera-based wildlife monitoring is often overwhelmed by non-target triggers and slowed by manual review or cloud-dependent inference, which can prevent timely intervention for high stakes human–wildlife conflicts. Our key contribution is a deployable, fully offline edge vision sensor that achieves near-real-time, highly accurate wildlife event classification by combining detector-based empty-image suppression with a lightweight classifier trained with a staged transfer-learning curriculum. Specifically, Stage 1 uses a pretrained You Only Look Once (YOLO)-family detector for permissive animal localization and empty-trigger suppression, and Stage 2 uses a lightweight EfficientNet-based binary classifier to confirm puma on detector crops and gate downstream actions. Our design is robust to low-quality nighttime monochrome imagery (motion blur, low contrast, illumination artifacts, and partial-body captures) and operates using commercially available components in connectivity-limited settings. In field deployments running since May 2025, end-to-end latency from camera trigger to action command is approximately 4 s. Ablation studies using a dataset of labeled wildlife images (pumas, not pumas) show that the two-stage approach substantially reduces false alarms in identifying pumas relative to a full-image classifier while maintaining high recall. On the held-out test set (N = 1434 events), the proposed two-stage cascade achieves precision 0.983, recall 0.975, F1 0.979, accuracy 0.986, and balanced accuracy 0.983, with only 8 false positives and 12 false negatives. The system can be easily adapted for other species, as demonstrated by rapid retraining of the second stage to classify ringtails. Downstream responses (e.g., notifications and optional audio/light outputs) provide flexible actuation capabilities that can be configured to support intervention.

58 GEOSCIENCES↗
DOE OSTI2025

Energy Infrastructure Futures: A Multiscale Evaluation of Projected Power Plant Siting Across the Western Interconnection

Energy Infrastructure Futures: A Multiscale Evaluation of Projected Power Plant Siting Across the Western Interconnection Description This dataset contains input and output data for the manuscript Mongird, K. et al. (under review) titled "Energy Infrastructure Futures: A Multiscale Evaluation of Projected Power Plant Siting Across the Western Interconnection". Input data corresponds to gridded spatial siting attributes that are necessary to conduct a random forest machine learning analysis of siting feature importance. Output data includes SHAP feature analysis outputs, and classification report values. For data on power plant siting results referred to in the manuscript, please refer to the CERF: IM3 Projected Western US Power Plant Locations data download page. The downloadable data includes values for eight different future scenarios for the Western US. The scenarios include combinations of two Shared Socioeconomic Pathways (SSP3 and SSP5) with four high-resolution climate projections specific to the United States (see, https://tgw-data.msdlive.org/). These climate projections include "hotter" and "cooler" variants for two Representative Concentration Pathways (RCP4.5 and RCP8.5). The resulting eight simulations are: rcp45cooler_ssp3 rcp45cooler_ssp5 rcp45hotter_ssp3 rcp45hotter_ssp5 rcp85cooler_ssp3 rcp85cooler_ssp5 rcp85hotter_ssp3 rcp85hotter_ssp5 Technical Information The dataset includes two sets of data files: (1) CERF gridded siting parameters and (2) Feature analysis outputs and classification reports. All downloadable data is in csv file format. Files with x/y coordinate information use the Albers Equal Area Conic projection (ESRI:102003). 1. CERF Gridded Siting Parameters This directory provides a balanced sample of gridded CERF siting parameters data for eight different scenarios for the Western US through 2055, seven different technologies, and eight timesteps. This data serves as input to the feature analysis. It contains the following parameters. region_name - name of region (i.e., state) sited - binary value representing whether the grid cell received a siting of that technology type (1=True) rcp - binary value representing scenario resource concentration pathway (0 = RCP4.5, 1 = RCP8.5) ssp - binary value representing scenario shared socioeconomic pathway (0 = SSP3, 1 = SSP5) climate - binary value representing cooler (0) or hotter (1) GCM forcing tech_name - generation technology name sited_year - year that values correspond to transmission_cost - cost of transmission interconnection pipeline_cost - cost of natural gas pipeline interconnection interconnection_cost - total interconnection cost (sum of transmission cost and gas pipeline cost) lmp - associated locational marginal value ($/MWh) associated with the grid cell, timestep, scenario, and technology xcoord - x-coordinate of location ycoord - y-coordinate of location 2a. Feature Analysis Output The dataset includes the feature analysis shap output for locational marginal price and interconnection cost. It contains the following parameters. technology - generator technology name scenario - name of scenario feature - name of feature, either locational_marginal_price or interconnection_cost value - the mean of absolute value of SHAP values for given feature 2b. Feature Analysis Classification Report This download includes the classification report associated with each random forest model. The dataset contains the following parameters. technology - generation technology name scenario - name of scenario test - one of precision (the proportion of predicted positives that are actually correct), recall (the proportion of actual positives that were correctly identified), f1-score (the harmonic mean of precision and recall) 0.0 - value of test for classification of 0 (grid cell not chosen for siting) 1.0 - value of test for classification of 1 (grid cell chosen for siting) accuracy - accuracy of model (i.e., fraction of all predictions that were right) macro avg - Simple average of test values for all classes weighted avg - Weighted average of test values for all classes, weighted based on Acknowledgment IM3 is a multi-institutional effort led by Pacific Northwest National Laboratory and supported by the U.S. Department of Energy's Office of Science as part of research in MultiSector Dynamics, Earth and Environmental Systems Modeling Program. License This data is made available under a CCBY4 License Disclaimer This material was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor the United States Department of Energy, nor the Contractor, nor any or their employees, nor any jurisdiction or organization that has cooperated in the development of these materials, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness or any information, apparatus, product, software, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or Battelle Memorial Institute. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. PACIFIC NORTHWEST NATIONAL LABORATORYoperated byBATTELLEfor theUNITED STATES DEPARTMENT OF ENERGYunder Contract DE-AC05-76RL01830

Mongird, Kendall [Pacific Northwest National Labor↗
DOE OSTI2024

ML Classifier Fusion for Three Data Streams with Quality Inversely Proportional to Time Resolution

We consider a monitoring scenario of phenomenon using three different streams of measurements whose quality is proportional to their constant inter-arrival times. Each measurement of a stream needs to be binary-classified to reflect the state of interest of the phenomenon. A set of classifiers is separately trained and fused for each stream at its time resolution using measurements collected under known states. We present a machine learning method to fuse the outputs of these fusers to provide a final classification at the finest time resolution. We show that this fused-fusers method provides decisions with likely superior classification probability compared to the best individual classifiers and fused-classifiers. We derive generalization equations that guarantee a superior classification probability of fused-fusers with a confidence probability specified by the classifiers’ generalization equations. We apply these results to study a practical problem of classifying Pu/Np target dissolution events at a radiochemical processing facility using gamma spectral measurements of effluent flows.

Rao, Nageswara↗
DOE OSTI2024

Neural units with time-dependent functionality

We show that the time-resolved dynamics of an underdamped harmonic oscillator can be used to do multifunctional computation, performing distinct computations at distinct times within a single dynamical trajectory. We consider the amplitude of an oscillator whose inputs influence its frequency. The activity of the oscillator at fixed times is a nonmonotonic function of its inputs, so it can solve problems such as XOR that are not linearly separable. The activity of the oscillator at fixed input is a nonmonotonic function of time, so it is multifunctional in a temporal sense, and able to carry out distinct nonlinear computations at distinct times within the same dynamical trajectory. We show that a single oscillator, observed at different times, can act as all of the elementary logic gates and perform binary addition, the latter usually implemented in hardware using five logic gates. We show that a set of n oscillators, observed at different times, can perform an arbitrary number of analog-to-n-bit digital conversions. We also show that oscillators can be trained by gradient descent to perform distinct classification tasks at distinct times. Computing with time-dependent functionality can be done in or out of equilibrium, and suggests a way of reducing the number of parameters or devices required to do nonlinear computations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Physics-inspired spatiotemporal-graph AI ensemble for the detection of higher order wave mode signals of spinning binary black hole mergers

We present a new class of AI models for the detection of quasi-circular, spinning, non-precessing binary black hole mergers whose waveforms include the higher order gravitational wave modes ($\ell$, |m|) = {(2,2), (2,1), (3,3), (3,2), (4,4)}, and mode mixing effects in the $\ell$ = 3, |m| = 2 harmonics. These AI models combine hybrid dilated convolution neural networks to accurately model both short- and long-range temporal sequential information of gravitational waves; and graph neural networks to capture spatial correlations among gravitational wave observatories to consistently describe and identify the presence of a signal in a three detector network encompassing the Advanced LIGO and Virgo detectors. We first trained these spatiotemporal-graph AI models using synthetic noise, using 1.2 million modeled waveforms to densely sample this signal manifold, within 1.7 h using 256 NVIDIA A100 GPUs in the Polaris supercomputer at the Argonne Leadership Computing Facility. This distributed training approach exhibited optimal classification performance, and strong scaling up to 512 NVIDIA A100 GPUs. With these AI ensembles we processed data from a three detector network, and found that an ensemble of 4 AI models achieves state-of-the-art performance for signal detection, and reports two misclassifications for every decade of searched data. We distributed AI inference over 128 GPUs in the Polaris supercomputer and 128 nodes in the Theta supercomputer, and completed the processing of a decade of gravitational wave data from a three detector network within 3.5 h. Finally, we fine-tuned these AI ensembles to process the entire month of February 2020, which is part of the O3b LIGO/Virgo observation run, and found 6 gravitational waves, concurrently identified in Advanced LIGO and Advanced Virgo data, and zero false positives. This analysis was completed in one hour using one NVIDIA A100 GPU.

79 ASTRONOMY AND ASTROPHYSICS↗