Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep data”

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 1,279 records · Page 71
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Deep Dive Topic: Choosing between ablators

Recent data on implosions using identical hohlraums and very similar laser drives underscores the conundrum of making a clear choice of one ablator over another. Table I shows a comparison of Be and CH in a nominal length, gold, 575 μm-diameter, 1.6 mg/cc He gas-fill hohlraum while Table II shows a comparison of undoped HDC and CH in a +700 length, gold, 575 μm diameter, 1.6 mg/cc He gas fill hohlraum. As can be seen in the tables, the net integrated fusion performance of these ablators is the same to within error bars. In the case of the undoped HDC and CH ablators, the hot spot shapes of the implosions were nearly indistinguishable for the experiments listed in Table II.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Hypothesis testing via AI: Generating physically interpretable models of scientific data with machine learning (Full Technical Report)

Deep learning has demonstrated an exceptional ability to solve complex tasks (an engineering success); however, it has done so at the expense of the ability to generate new knowledge (a scientific failure). We propose an alternative framework—entitled Deep Symbolic Regression (DSR)—in which artificial neural networks (NNs) rapidly generate hypotheses about physical relationships among inputs. This framework bypasses the need to interpret an NN altogether, while still leveraging the representational power of deep learning. The resulting models are tractable mathematical expressions, which are inherently and readily human interpretable and can provide insights into underlying physical phenomena. Further, we fold this methodology into the scientific process by allowing the scientist to directly integrate a priori knowledge and beliefs to accelerate learning. We demonstrate this methodology on symbolic regression—the problem of rediscovering underlying expressions describing a dataset—and achieve state-of-the-art performance across a wide variety of symbolic regression problems. Further, we generalize our DSR framework to apply to the more general class of symbolic optimization problems, in which one seeks to optimize a sequence of symbols or “tokens” under a black-box reward function. Examples of other symbolic optimization problems include neural architecture search and computational antibody design. Our generalized tool, Deep Symbolic Optimization (DSO), has been demonstrated on the task of learning symbolic control policies for reinforcement learning environments, and has been adopted as an enabling capability for computational antibody design.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2022

Global Reduction in Ship-tracks from Sulfur Regulations for Shipping Fuel

Ship-tracks are produced by ship-emitted aerosols interacting with low clouds. Here, we apply deep learning models on satellite data to produce the first global climatology map of ship-tracks. We show that ship-tracks are at the nexus of cloud physics, maritime shipping, and fuel regulation. Our map captures major shipping lanes while missing others because of background conditions. Ship-track frequency is more than 10 times higher than a previous survey, and its interannual fluctuations reflect variations in cross-ocean trade, shipping activity, and fuel regulations. Fuel regulation can alter both detected frequency and shipping routes due to cost. The 2020 fuel regulation, together with the coronavirus disease 2019 pandemic, reduced ship-track frequency to its lowest level in recent decades across the globe and may have ushered in an era of low frequency. The regulation reduces the aerosol indirect forcing from ship emissions by 46% or between 0.02 and 0.27 W m −2 given its current estimates.

ship tracks↗
NASA NTRS2014

Advancing Navigation, Timing, and Science with the Deep Space Atomic Clock

NASA's Deep Space Atomic Clock mission is developing a small, highly stable mercury ion atomic clock with an Allan deviation of at most 1e-14 at one day, and with current estimates near 3e-15. This stability enables one-way radiometric tracking data with accuracy equivalent to and, in certain conditions, better than current two-way deep space tracking data; allowing a shift to a more efficient and flexible one-way deep space navigation architecture. DSAC-enabled one-way tracking will benefit navigation and radio science by increasing the quantity and quality of tracking data. Additionally, DSAC would be a key component to fully-autonomous onboard radio navigation useful for time-sensitive situations. Potential deep space applications of DSAC are presented, including orbit determination of a Mars orbiter and gravity science on a Europa flyby mission.

radio science↗
NASA NTRS1978

Helios mission support

Activities of the Deep Space Network Operations organization in support of the Helios Project from 15 October 1977 through 15 December 1977 are described. Topics covered include: (1) Mark 3 data subsystem testing at the conjoint Deep Space Stations (DSS) 42/43 (Canberra, Australia); (2) MDS implementation at DSS 61/63 (Madrid, Spain); (3) Radio Science update, and (4) other mission-related activities.

Goodwin, P. S.↗
NASA NTRSDate not supplied

Increasing accessibility to deep learning-based analytics for space biology: pretrained models, transfer learning, and analytics platform development

Biological systems react in complex ways to the stressors of spaceflight, and the data capturing these relationships is concomitantly high-dimensional and complex. Deep learning and machine learning approaches are increasingly popular as an analytical approach for space biosciences, due to their ability to model complex relationships in complex data. However, such approaches often require large datasets and extensive computational resources. New approaches that minimize data sizes and computational power needed to leverage machine learning, and resources that make these approaches accessible, are needed to increase accessibility and adoption of machine learning in the space biosciences. Transfer learning, in which a pretrained model of broad utility is trained on a large dataset, and subsequently reused on downstream applications for which data is more limited, is one approach to minimizing data and computational intensity of deep learning applications. This transfer learning approach results in more performant models in high-dimensional, low-sample-size settings such as space biology, as compared to training models on limited data from scratch. This presentation will outline efforts to generate pretrained models for the space biology community, and highlight transfer learning applications modeling microbial antibiotic resistance during spaceflight. Finally, in order to increase accessibility of these models and tools, as well as others, for the broader space biology community, we present a modeling and analysis platform facilitating machine learning applications in space biology. This platform streamlines machine learning training and analysis in a notebook format, facilitates download and use of space biology data from the NASA GeneLab database, and can be utilized on NASA-hosted servers or downloaded and hosted locally. This effort, as part of the AI4LS (Artificial Intelligence for Life in Space) working group, will increase accessibility, feasibility, and performance of machine learning approaches for the space biology community.

Adrienne Hoarfrost↗
DOE OSTI2023

Scaling Subspace-Driven Approaches Using Information Fusion

In this work, we seek to exploit the deep structure of multi-modal data to robustly exploit the group subspace distribution of the information using the Convolutional Neural Networks (CNNs) formalism. Upon unfolding the set of subspaces constituting each data modality, and learning their corresponding encoders, an optimized integration of the generated inherent information is carried out to yield a characterization of various classes. Referred to as deep Multimodal Robust Group Subspace Clustering (DRoGSuRe), this approach is compared against the independently developed state-of-the-art approach named Deep Multimodal Subspace Clustering (DMSC). Experiments on different multimodal datasets show that our approach is competitive and more robust in the presence of noise.

Ghanem, Sally↗
DOE OSTI2017

Deep learning for computational chemistry

The rise and fall of artificial neural networks is well documented in the scientific literature of both the fields of computer science and computational chemistry. Yet almost two decades later, we are now seeing a resurgence of interest in deep learning, a machine learning algorithm based on “deep” neural networks. Within the last few years, we have seen the transformative impact of deep learning the computer science domain, notably in speech recognition and computer vision, to the extent that the majority of practitioners in those field are now regularly eschewing prior established models in favor of deep learning models. In this review, we provide an introductory overview into the theory of deep neural networks and their unique properties as compared to traditional machine learning algorithms used in cheminformatics. By providing an overview of the variety of emerging applications of deep neural networks, we highlight its ubiquity and broad applicability to a wide range of challenges in the field, including QSAR, virtual screening, protein structure modeling, QM calculations, materials synthesis and property prediction. In reviewing the performance of deep neural networks, we observed a consistent outperformance against non neural networks state-of-the-art models across disparate research topics, and deep neural network based models often exceeded the “glass ceiling” expectations of their respective tasks. Coupled with the maturity of GPU-accelerated computing for training deep neural networks and the exponential growth of chemical data on which to train these networks on, we anticipate that deep learning algorithms will be a useful tool and may grow into a pivotal role for various challenges in the computational chemistry field.

Computational Chemistry↗
DOE OSTI2025

Satellite Imagery of PV Site Storm Damage

"This repository contains multiple data sets focused on visible damage to photovoltaic (PV) installations following extreme weather events such as hailstorms and hurricanes. Data sets are split into two categories: the first category, the ‘manually labeled’ data, was compiled by researchers manually, and contains manually identified PV sites exposed to storms. The second data set, the ‘aggregated’ data, is a compilation of the manually labeled PV sites and deep learning-identified PV sites. The hail damage data set focuses on post-storm PV damage following a September 24, 2023 hailstorm in Austin, TX, which caused over $600 million in damages in the Austin metro area. The hurricane damage data set focuses on post-storm PV damage following Hurricanes Irma and Maria in Puerto Rico and the US Virgin Islands. Hurricanes Irma and Maria were back-to-back category 5 hurricanes, which pummeled the Caribbean and southeastern United States in September 2017, causing an estimated $115.2 billion in damages."

14 SOLAR ENERGY↗
DOE OSTI2025

Predicting critical heat flux with uncertainty quantification and domain generalization using conditional variational autoencoders and deep neural networks

Deep generative models (DGMs) can generate synthetic data samples that closely resemble the original dataset, addressing data scarcity. In this work, we developed a conditional variational autoencoder (CVAE) to augment critical heat flux (CHF) data used for the 2006 Groeneveld lookup table. To compare with traditional methods, a fine-tuned deep neural network (DNN) regression model was evaluated on the same dataset. Both models achieved small mean absolute relative errors, with the CVAE showing more favorable results. Uncertainty quantification (UQ) was performed using repeated CVAE sampling and DNN ensembling. The DNN ensemble improved performance over the baseline, while the CVAE maintained consistent results with less variability and higher confidence. Both models achieved small errors inside and outside the training domain, with slightly larger errors outside. Altogether, the CVAE performed better than the DNN in predicting CHF and exhibited better uncertainty behavior.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Knowledge-guided learning with curated prior genetic biomarkers for robust model interpretation

Abstract Motivation Knowledge-guided learning offers effective and robust model training strategies in data-scarce settings by incorporating established domain knowledge, thereby enhancing generalization, robustness, and interpretability. By contrast, conventional deep learning approaches rely purely on data-driven learning, which can limit robust model interpretability, particularly in high-dimensional settings with limited size samples. In computational biology, knowledge-guided learning has primarily leveraged network- and structural-based knowledge, leading to biologically interpretable representations and enhanced predictive performance compared to conventional approaches. However, curated biomarkers, one of the most accessible forms of biological knowledge, remain largely unexplored within knowledge-guided paradigms. Results In this study, we propose a model-agnostic training paradigm, Biomarker-driven Explainable Prior-guided Learning (BioExPL), that can be applied to any neural networks that incorporates curated prior knowledge. BioExPL enforces neural networks to reflect curated biomarker priors in their latent representations through a novel knowledge-alignment loss. BioExPL consistently demonstrated significantly improved predictive performance and enhanced model interpretability with minimized computational overhead in simulation studies and intensive experiments on multiple cancer datasets. BioExPL not only integrates prior curated knowledge into the model but also accurately identifies unknown associated signals additionally. BioExPL is model-agnostic and domain-independent, enabling its integration into diverse neural network architectures. Availability and implementation The open-source is publicly available at: https://github.com/datax-lab/BioExPL.

Baek, Beomsu [Department of Computer Science, Univ↗
DOE OSTI2025

Synthetic Infrasound Data for Machine Learning Detectors

Synthetic data is a powerful tool to generate large amounts of training data for machine learning models. The methods outlined in this report will be used to retrain the deep learning classifier for increased accuracy. Synthetic data will be useful to address the natural class imbalance between the different categories in the original ML work. Additionally, these tools will be applied for a variety of signal analysis methods that would use signals with a known signal-to-noise ratio for validation and testing.

58 GEOSCIENCES↗
NASA NTRS1987

Recurring polynyas over the Cosmonaut Sea and the Maud Rise

Two remarkable deep ocean polynyas observed in the Antarctic region during the winter of 1980, here referred to as the Cosmonaut polynya and the Maud Rise polynya, are discussed. It is proposed that both polynyas are products of deep-reaching convection which introduces warmer deep water into the surface layer. Hydrographic data at both sites indicate the existence of localized doming of the pycnocline. This brings warmer, saltier deep water close to the sea surface, which has been demonstrated to be an effective preconditioner for deep-reaching convection. A possible relationship between the two polynyas is sugggested in that both are in the eastern margins of the Weddell Sea.

Comiso, J. C.↗
NASA NTRS2023

Damage Detection of a Pressure Vessel with Smart Sensing and Deep Learning

Structural Health Monitoring plays a crucial role in ensuring the safety and reliability of critical infrastructure, including pressure vessels involved in various applications. This research reports the damage detection of a pressure box employed in space habitat that operates in harsh environment where both structural failure and bolt joint loosening may occur. These failure modes are extremely hard to model based on first principles. We explore proper sensing mechanism and the associated inverse analysis algorithm that can elucidate the health condition of the pressure box. It is identified that piezoelectric impedance based active interrogation can provide necessary information for damage detection in such a system. Concurrently, deep learning technique leveraging spatial convolutional neural network is synthesized to analyze the raw data acquired and identify different types of damage. By training the deep learning model on a dataset of healthy and various damage scenarios, we can achieve high accuracy in identifying the presence of damage and its type. This research provides a data-driven methodology for structural damage detection using deep learning and has the potential to be extended to various systems with different failure modes.

Yang Zhang↗
DOE OSTI2023

Riverbed and Near-Surface Water Quality Data, Hanford Reach, Columbia River, February 2021 - April 2022

This dataset contains longitudinal profiles of natural groundwater tracers (temperature, electrical conductivity (EC), and Radon-222 (Rn)) collected to identify locations of hydrologic exchange flows along a 75-km reach of the Columbia River near Richland, Washington.A jetboat was used to tow a 30 foot long, weighted tether that had one set of temperature and EC sensors just below the water surface and one set of temperature and EC sensors on the end (just above the river bed). The dataset contains profiles of temperature/EC data (resolution: 1 second) collected along the left and right banks during three different sampling events. During the final sampling event, we revisited some "deep holes" where the river was more than 30 feet deep with a weighted tether that was 75 feet long. The dataset has been processed to remove data collected when the sensors were not in the water, or when the water was too deep to keep the tether on the riverbed. We collected grab samples of water for analysis of dissolved Rn in places where the temperature/EC was different from the background river water. The Rn data was joined to the temperature/EC profiles based on the sample times.This dataset is comprised of one main data folder containing (1) file-level metadata; (2) data dictionary; (3) installation methods; (4) February 2021 data; (5) July 2021 data; (6) April 2022 data; and (7) April 2022 deep hole data. All files are .csv format and can be opened with a spreadsheet program (such as Microsoft Excel or OpenOffice).

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2000

Fossil Record of Precambrian Life on Land

The argument that the earth's early ocean was up to two times modern salinity was published in 'Nature' and presented at the 1998 Annual Meeting of the Geological Society of America in Toronto. The argument is bolstered by chemical data for fluid inclusions in Archean black smokers. The inclusions were 1.7 times the modern salinity causing the authors to interpret the parent fluids as evaporite brines (in a deep marine setting). I reinterpreted the data in terms of the predicted value of high Archean salinities. If the arguments I presented are on track, early life was either halophilic or non-marine. Halophiles are not among the most primitive organisms based on RNA sequencing, so here is an a priori argument that non-marine environments may have been the site of most early biologic evolution. This result carries significant implications for the issue of past life on Mars or current life on the putative sub-ice oceans on Europa and possibly Callisto. If the Cl/H 2 O ratio on these objects is similar to that of the earth, then oceans and oceanic sediments are probably not the preferred sites for early life. On Mars, this means that non-marine deposits such as caliche in basalt may be an overlooked potential sample target.

Paul Knauth↗