SEARCH · Search NASA

Results for “deep learning benchmarks”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3

All sources NASA NTRS DOE OSTI

DOE OSTI2026

scPlantAnnotate: an accurate and robust transformer-based model for plant cell type annotation

Accurate cell type annotation remains a major bottleneck in plant single-cell RNA sequencing (scRNA-seq), where existing tools are often adapted from animal studies and perform sub-optimally on plant data. The lack of plant-specific computational frameworks limits the construction of plant cell atlases and downstream biological discovery. We develop and evaluate scPlantAnnotate, a Transformer-based reference annotation framework tailored for plant scRNA-seq data, and benchmark it against state-of-the-art deep learning and conventional methods across multiple plant species. Species-specific scPlantAnnotate models were trained using curated datasets from Arabidopsis thaliana, Zea mays, Oryza sativa, and Glycine max. We compared scPlantAnnotate with leading baselines under both standard random-split evaluation and a more stringent leave-one-dataset-out setting, which tests robustness to completely unseen datasets and tissue types. scPlantAnnotate consistently outperforms existing approaches across all four species under random-split evaluation. In the leave-one-dataset-out setting for A. thaliana, where performance drops markedly for all methods due to strong batch effects and dataset heterogeneity, scPlantAnnotate nonetheless achieves the highest Accuracy, Macro-F1, Balanced Accuracy, and Macro-AUROC on average and ranks first on most held-out datasets. These results demonstrate improved robustness to dataset shifts, a critical yet underexplored challenge in plant scRNA-seq analysis. A freely accessible web server enables users to annotate their own datasets using pretrained models. scPlantAnnotate provides a plant-specific, Transformer-based framework for single-cell annotation that delivers state-of-the-art performance and enhanced robustness to unseen datasets. By addressing limitations of existing tools and enabling scalable reference-based annotation, scPlantAnnotate supports the development of comprehensive plant cell atlases and facilitates broader use of single-cell genomics in plant biology.

Bioinformatics↗

DOE OSTI2020

A Molecular-MNIST Dataset for Machine Learning Study on Diffraction Imaging and Microscopy

An image dataset of 10 different size molecules, where each molecule has 2,000 structural variants, is generated from the 2D cross-sectional projection of Molecular Dynamics trajectories. The purpose of this dataset is to provide a benchmark dataset for the increasing need of machine learning, deep learning and image processing on the study of scattering, imaging and microscopy.

BASIC BIOLOGICAL SCIENCES,OTHER INSTRUMENTATION↗

NASA NTRS2020

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-ofthe-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Hesham ElAbd↗

DOE OSTI2020

Using Neural Architecture Search for Improving Software Flaw Detection in Multimodal Deep Learning Models

Software flaw detection using multimodal deep learning models has been demonstrated as a very competitive approach on benchmark problems. In this work, we demonstrate that even better performance can be achieved using neural architecture search (NAS) combined with multimodal learning models. We adapt a NAS framework aimed at investigating image classification to the problem of software flaw detection and demonstrate improved results on the Juliet Test Suite, a popular benchmarking data set for measuring performance of machine learning models in this problem domain.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Comprehensive assessment of deep reinforcement learning approaches for economic dispatch in nuclear-driven microgrids

As the electrical grid integrates more variable renewable energy sources such as wind and solar, the demand for distributed and flexible systems to address this increased variability becomes critical. Nuclear-driven microgrids provide a promising solution by offering stable generation to complement intermittent renewables, ensuring grid reliability and operating efficiency. This paper proposes a recurrent deep reinforcement learning framework for optimal economic dispatch in a nuclear-powered microgrid integrating renewable energy sources, small modular reactors, battery storage systems, and balance-of-plant dynamics. A three-agent control architecture is developed, where demand and renewable energy agents act as forecasters, and a reinforcement learning-based dispatch agent performs real-time energy allocation. A nonlinear programming formulation is first used to generate an optimal baseline for benchmarking. The proposed dispatch controller, based on Proximal Policy Optimization enhanced with Long Short-Term Memory networks, exploits temporal correlations in system dynamics by taking advantage of the time series used as inputs to improve policy robustness under uncertainty. Comparative analysis against established deep reinforcement learning methods, including Proximal Policy Optimization with a feedforward architecture, Soft Actor-Critic, and Twin Delayed Deep Deterministic Policy Gradient, demonstrates superior performance. Numerical results indicate that the proposed controller achieves a 0.39% cost reduction relative to the nonlinear programming benchmark and outperforms other learning-based methods by generating additional revenue of up to 0.35%. All reinforcement learning controllers compute dispatch actions in less than 0.3 s, resulting in a computational speedup of more than three orders of magnitude over the nonlinear programming baseline. The findings of this paper highlight their applicability for real-time operation and control in nuclear-integrated microgrids under volatile operating conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2023

Physics-Informed Deep Learning for Reconstruction of Spatial Missing Climate Information in the Antarctic

Understanding the influence of the Antarctic on the global climate is crucial for the prediction of global warming. However, due to very few observation sites, it is difficult to reconstruct the rational spatial pattern by filling in the missing values from the limited site observations. To tackle this challenge, regional spatial gap-filling methods, such as Kriging and inverse distance weighted (IDW), are regularly used in geoscience. Nevertheless, the reconstructing credibility of these methods is undesirable when the spatial structure has massive missing pieces. Inspired by image inpainting, we propose a novel deep learning method that demonstrates a good effect by embedding the physics-aware initialization of deep learning methods for rapid learning and capturing the spatial dependence for the high-fidelity imputation of missing areas. We create the benchmark dataset that artificially masks the Antarctic region with ratios of 30%, 50% and 70%. The reconstructing monthly mean surface temperature using the deep learning image inpainting method RFR (Recurrent Feature Reasoning) exhibits an average of 63% and 71% improvement of accuracy over Kriging and IDW under different missing rates. With regard to wind speed, there are still 36% and 50% improvements. In particular, the achieved improvement is even better for the larger missing ratio, such as under the 70% missing rate, where the accuracy of RFR is 68% and 74% higher than Kriging and IDW for temperature and also 38% and 46% higher for wind speed. In addition, the PI-RFR (Physics-Informed Recurrent Feature Reasoning) method we proposed is initialized using the spatial pattern data simulated by the numerical climate model instead of the unified average. Compared with RFR, PI-RFR has an average accuracy improvement of 10% for temperature and 9% for wind speed. When applied to reconstruct the spatial pattern based on the Antarctic site observations, where the missing rate is over 90%, the proposed method exhibits more spatial characteristics than Kriging and IDW.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Denoising Seismograms in the Time Domain Using a Deep Learning Model

Deep learning has emerged as a transformative tool for enhancing the extraction of reliable information from seismograms, addressing the increasing demand for precise and efficient seismic data analysis. We introduce an innovative encoder–decoder deep learning model, named WaveDenoiser, designed for noise reduction in the time domain, thereby eliminating the need for spectrogram computations that have been used for existing deep learning tools and significantly improving processing speed. Utilizing the benchmark dataset that is Stanford Earthquake Dataset, we developed three models of varying sizes: base, medium, and large. Notably, the large (referred to as WaveDenoiser) model demonstrated superior performance, achieving a median signal‐to‐noise ratio improvement of 8.8 dB on in‐distribution unseen data (in the same geographic region) and 7.7 dB on out‐distribution unseen data (in a new geographic region), outpacing both the base and medium models. Further evaluation of the WaveDenoiser model revealed a reduction in median arrival‐time errors by 0.02 s for P waves and 0.01 s for S waves when processing waveforms prior to phase picking using PhaseNet on in‐distribution unseen data. When tested on out‐distribution unseen data, the model also effectively reduced the P‐wave median arrival‐time error by 0.02 and 0.01 s in median arrival‐time error for S waves. Importantly, the application of WaveDenoiser resulted in a significant reduction of phase picking outliers by 1.1% to 3.6% for both P and S waves. In addition, we achieved over five times acceleration in processing speed compared with the seisBench implementation of DeepDenoiser. Our findings underscore the potential of WaveDenoiser as a powerful tool for improving seismic data analysis and processing efficiency.

P-waves↗

DOE OSTI2022

A benchmark dataset for Hydrogen Combustion

The generation of reference data for deep learning models is challenging for reactive systems, and more so for combustion reactions due to the extreme conditions that create radical species and alternative spin states during the combustion process. Here, we extend intrinsic reaction coordinate (IRC) calculations with ab initio MD simulations and normal mode displacement calculations to more extensively cover the potential energy surface for 19 reaction channels for hydrogen combustion. A total of ~290,000 potential energies and ~1,270,000 nuclear force vectors are evaluated with a high quality range-separated hybrid density functional, ωB97X-V, to construct the reference data set, including transition state ensembles, for the deep learning models to study hydrogen combustion reaction.

08 HYDROGEN↗

DOE OSTI2025

Postearthquake Damage Mapping via Remote Sensing: Lessons From the 2023 Türkiye Disaster

This review addresses the urgent need for scalable, accurate, and reproducible remote sensing solutions following the February 2023 Türkiye earthquakes. It synthesizes the contributions of five peer-reviewed studies published in the IEEE JSTARS Special Issue on postearthquake damage and risk assessment. These studies cover areas such as damage classification with deep learning, fusion of multisource remote sensing data, creation of benchmark datasets, detailed damage mapping, and analysis of geophysical signals using outgoing longwave radiation. The article summarizes the methodological approaches and the practical relevance of the reviewed studies for detecting, evaluating, and quantifying damage, and outlines key challenges, including model generalization, class ambiguity, and data integration. It also discusses emerging trends, including explainable artificial intelligence, multimodal data fusion, and open-data platforms. This synthesis provides a foundation for building robust, interpretable, and real-time disaster response systems and aims to guide future research in earthquake-related Earth observation and rapid damage assessment.

Taskin, Gulsen [Istanbul Technical University] (OR↗

DOE OSTI2023

Review of machine learning and deep learning models for toxicity prediction

The ever-increasing number of chemicals has raised public concerns due to their adverse effects on human health and the environment. To protect public health and the environment, it is critical to assess the toxicity of these chemicals. Traditional in vitro and in vivo toxicity assays are complicated, costly, and time-consuming and may face ethical issues. These constraints raise the need for alternative methods for assessing the toxicity of chemicals. Recently, due to the advancement of machine learning algorithms and the increase in computational power, many toxicity prediction models have been developed using various machine learning and deep learning algorithms such as support vector machine, random forest, k-nearest neighbors, ensemble learning, and deep neural network. This review summarizes the machine learning- and deep learning-based toxicity prediction models developed in recent years. Support vector machine and random forest are the most popular machine learning algorithms, and hepatotoxicity, cardiotoxicity, and carcinogenicity are the frequently modeled toxicity endpoints in predictive toxicology. It is known that datasets impact model performance. The quality of datasets used in the development of toxicity prediction models using machine learning and deep learning is vital to the performance of the developed models. The different toxicity assignments for the same chemicals among different datasets of the same type of toxicity have been observed, indicating benchmarking datasets is needed for developing reliable toxicity prediction models using machine learning and deep learning algorithms. This review provides insights into current machine learning models in predictive toxicology, which are expected to promote the development and application of toxicity prediction models in the future.

Research & Experimental Medicine↗

DOE OSTI2023

Demonstration of object location, classification, and characterization by developed deep learning dust ablation trail analysis code package using plasma jets

Based on deep learning, a Dust Ablation Trail Analysis (DATA) code package is developed to detect dust ablation trails in tokamaks, which is intended to analyze a large amount data of tokamak dusts. To validate and benchmark the DATA code package, 2440 plasma jet images are exploited for the training and test of the deep learning DATA code package, since plasma jets resemble the shape and size of dust ablation clouds in tokamaks. After being trained by 1920 plasma jet images, the DATA code package is able to locate 100% plasma jets, classify plasma jets with the accuracy of >99.9%, and output image skeleton information for classified plasma jets. The DATA code package trained by the plasma jet images is also used to analyze the dust ablation trails captured in the Experimental Advanced Superconducting (EAST) tokamak with the satisfactory performance, further verifying its applicability in the fusion dust ablation investigation. Based on its excellent performance presented here, it is demonstrated that our DATA code package is able to automatically identify and analyze dust ablation trails in tokamaks, which can be used for further detailed investigations, such as the three-dimensional reconstruction of dusts and their ablation trails.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2022

BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset

The BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset adds node-level energy consumption data from watt-meters to the primary sweep of the BUTTER - Empirical Deep Learning Dataset. This dataset contains energy consumption and performance data from 63,527 individual experimental runs spanning 30,582 distinct configurations: 13 datasets, 20 sizes (number of trainable parameters), 8 network "shapes", and 14 depths on both CPU and GPU hardware collected using node-level watt-meters. This dataset reveals the complex relationship between dataset size, network structure, and energy use, and highlights the impact of cache effects. BUTTER-E is intended to be joined with the BUTTER dataset (see "BUTTER - Empirical Deep Learning Dataset on OEDI" resource below) which characterizes the performance of 483k distinct fully connected neural networks but does not include energy measurements.

Array↗

DOE OSTI2020

Ensemble transfer learning for the prediction of anti-cancer drug response

Abstract Transfer learning, which transfers patterns learned on a source dataset to a related target dataset for constructing prediction models, has been shown effective in many applications. In this paper, we investigate whether transfer learning can be used to improve the performance of anti-cancer drug response prediction models. Previous transfer learning studies for drug response prediction focused on building models to predict the response of tumor cells to a specific drug treatment. We target the more challenging task of building general prediction models that can make predictions for both new tumor cells and new drugs. Uniquely, we investigate the power of transfer learning for three drug response prediction applications including drug repurposing, precision oncology, and new drug development, through different data partition schemes in cross-validation. We extend the classic transfer learning framework through ensemble and demonstrate its general utility with three representative prediction algorithms including a gradient boosting model and two deep neural networks. The ensemble transfer learning framework is tested on benchmark in vitro drug screening datasets. The results demonstrate that our framework broadly improves the prediction performance in all three drug response prediction applications with all three prediction algorithms.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2025

DOE FAIR Surrogate Benchmarks Supporting AI and Simulation Research (SBI Surrogate Benchmark Initiative) (Final Report)

Computational Science is being revolutionized by integrating AI and simulation and, in particular, by deep learning surrogate models that can replace all or part of traditional large‐scale HPC computations. Such surrogates can achieve remarkable performance improvements, as much as several orders of magnitude, and save both compute time and energy. The Surrogate Benchmark Initiative (SBI) project creates a community repository and FAIR (Findable, Accessible, Interoperable, and Reusable) data ecosystem for HPC application surrogate benchmarks. The SBI team comes from Argonne National Laboratory (ANL), Indiana University (IU), Rutgers University, the University of Tennessee, Knoxville (UTK), and the University of Virginia(UVA). SBI repositories include data, code, and all relevant collateral artifacts, that the science and engineering community needs to use and reuse these data sets and surrogates. SBI repositories generate active research from both participants in SBI and the broader AI and domain science communities. This project develops surrogates that use several different neural nets to learn and quickly infer the results of simulations and data systems and capture them as surrogate benchmarks with a rich set of metadata, covering. Data; Model; Metrics specification; Machine specification; Science, Speed, Power Results, We research FAIR metadata for these benchmarks. We develop application surrogate examples as benchmarks across many fields (ANL, UTK, IU, UVA). We also study non Surrogate benchmarks that have many common features and similar issues regarding FAIRness. We work with MLCommons (UVA, UTK), which is a major machine learning benchmarking activity where we get metadata ontologies, software, and benchmarks, benchmarks have datasets, models, and metadata, and they need a technical framework developed by UTK and Rutgers and deployed by UVA. We study features of Surrogates, including performance, training set size, and uncertainty quantification (Rutgers, UVA and IU).

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Load Profile Inpainting for Missing Load Data Restoration and Baseline Estimation

This paper introduces a Generative Adversarial Nets (GAN) based, Load Profile Inpainting Network (Load-PIN) for restoring missing load data segments and estimating the baseline for a demand response event. The inputs are time series load data before and after the inpainting period together with explanatory variables (e.g., weather data). Here, we propose a Generator structure consisting of a coarse network and a fine-tuning network. The coarse network provides an initial estimation of the data segment in the inpainting period. The fine-tuning network consists of self-attention blocks and gated convolution layers for adjusting the initial estimations. Loss functions are specially designed for the fine-tuning and the discriminator networks to enhance both the point-to-point accuracy and realisticness of the results. We test the Load-PIN on three real-world data sets for two applications: patching missing data and deriving baselines of conservation voltage reduction (CVR) events. We benchmark the performance of Load-PIN with five existing deep-learning methods. Our simulation results show that, compared with the state-of-the-art methods, Load-PIN can handle varying-length missing data events and achieve 15-30% accuracy improvement.

14 SOLAR ENERGY↗

DOE OSTI2022

EQ_phase_detection

The EQ_phase_detection software is designed to scan continuous daily waveforms to detect earthquake phase arrivals from local to regional (150 km) events. The detections are made with a deep learning encoder-decoder model. When the model detects an earthquake in the waveforms, a second model is implemented to classify the first arriving motions. Both deep learning models are trained with the Tensorflow package using publicly available benchmark data sets. The software input is a path to a directory that contains waveforms in mseed format and the associated response files in xml format. The output is a data table of time stamped detections, signal amplitude, signal-to-noise ratio, and softmax probability of the detection in a generic format applicable to post-processing association algorithms for event locations. Additionally, the p-wave and s-wave waveforms are saved in a data table for rapid access when producing improved locations using correlation-based techniques. The software is designed for multiprocessing with multiple GPU’s for rapid processing of large data sets. The configuration file provides flexibility in the trained models implemented and allows access to multiple models trained for different sampling rates or input dimensions. This is particularly useful for regions with multiple networks that do not have the same data parameters.

Johnson, Christopher↗

DOE OSTI2020

Using Neural Architecture Search for Improving Software Flaw Detection in Multimodal Deep Learning Models

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

Combination of deep neural network with attention mechanism enhances the explainability of protein contact prediction

Abstract Deep learning has emerged as a revolutionary technology for protein residue‐residue contact prediction since the 2012 CASP10 competition. Considerable advancements in the predictive power of the deep learning‐based contact predictions have been achieved since then. However, little effort has been put into interpreting the black‐box deep learning methods. Algorithms that can interpret the relationship between predicted contact maps and the internal mechanism of the deep learning architectures are needed to explore the essential components of contact inference and improve their explainability. In this study, we present an attention‐based convolutional neural network for protein contact prediction, which consists of two attention mechanism‐based modules: sequence attention and regional attention. Our benchmark results on the CASP13 free‐modeling targets demonstrate that the two attention modules added on top of existing typical deep learning models exhibit a complementary effect that contributes to prediction improvements. More importantly, the inclusion of the attention mechanism provides interpretable patterns that contain useful insights into the key fold‐determining residues in proteins. We expect the attention‐based model can provide a reliable and practically interpretable technique that helps break the current bottlenecks in explaining deep neural networks for contact prediction. The source code of our method is available at https://github.com/jianlin-cheng/InterpretContactMap .

59 BASIC BIOLOGICAL SCIENCES↗