Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “deep learning method”

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 1,081 records · Page 60
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

General-Purpose Bayesian Tensor Learning With Automatic Rank Determination and Uncertainty Quantification

A major challenge in many machine learning tasks is that the model expressive power depends on model size. Low-rank tensor methods are an efficient tool for handling the curse of dimensionality in many large-scale machine learning models. The major challenges in training a tensor learning model include how to process the high-volume data, how to determine the tensor rank automatically, and how to estimate the uncertainty of the results. While existing tensor learning focuses on a specific task, this paper proposes a generic Bayesian framework that can be employed to solve a broad class of tensor learning problems such as tensor completion, tensor regression, and tensorized neural networks. We develop a low-rank tensor prior for automatic rank determination in nonlinear problems. Our method is implemented with both stochastic gradient Hamiltonian Monte Carlo (SGHMC) and Stein Variational Gradient Descent (SVGD). We compare the automatic rank determination and uncertainty quantification of these two solvers. We demonstrate that our proposed method can determine the tensor rank automatically and can quantify the uncertainty of the obtained results. We validate our framework on tensor completion tasks and tensorized neural network training tasks.

Bayesian inference↗
DOE OSTI2021

Quantum-assisted associative adversarial network: applying quantum annealing in deep learning

Abstract Generative models have the capacity to model and generate new examples from a dataset and have an increasingly diverse set of applications driven by commercial and academic interest. In this work, we present an algorithm for learning a latent variable generative model via generative adversarial learning where the canonical uniform noise input is replaced by samples from a graphical model. This graphical model is learned by a Boltzmann machine which learns low-dimensional feature representation of data extracted by the discriminator. A quantum processor can be used to sample from the model to train the Boltzmann machine. This novel hybrid quantum-classical algorithm joins a growing family of algorithms that use a quantum processor sampling subroutine in deep learning, and provides a scalable framework to test the advantages of quantum-assisted learning. For the latent space model, fully connected, symmetric bipartite and Chimera graph topologies are compared on a reduced stochastically binarized MNIST dataset, for both classical and quantum sampling methods. The quantum-assisted associative adversarial network successfully learns a generative model of the MNIST dataset for all topologies. Evaluated using the Fréchet inception distance and inception score, the quantum and classical versions of the algorithm are found to have equivalent performance for learning an implicit generative model of the MNIST dataset. Classical sampling is used to demonstrate the algorithm on the LSUN bedrooms dataset, indicating scalability to larger and color datasets. Though the quantum processor used here is a quantum annealer, the algorithm is general enough such that any quantum processor, such as gate model quantum computers, may be substituted as a sampler.

Wilson, Max (ORCID:0000000207983391)↗
DOE OSTI2025

Uncovering multiscale structure-property correlations via active learning in scanning tunneling microscopy

Atomic arrangements and local sub-structures fundamentally influence emergent material functionalities. These structures are conventionally probed using spatially resolved studies and the property correlations are deciphered by a researcher based on sequential explorations, thereby limiting the efficiency and scope. Here we demonstrate a multi-scale Bayesian deep-learning based framework that automatically correlates material structure with its electronic properties using scanning tunneling microscopy (STM) measurements in real-time. Its predictions are used to autonomously direct exploration toward regions of the sample that optimize a given material property. This method is deployed on a low-temperature ultra-high vacuum STM to understand the structure-property relationship in a europium-based semimetal, EuZn 2 As 2 , a promising candidate relevant to magnetism-driven topological phenomena. The framework employs a sparse-sampling approach to efficiently construct the scalar-property space using minimal measurements, about 1–10% of the data required in standard hyperspectral methods. Moreover, we formulate the problem hierarchically across length scales, implementing autonomous workflow to locate mesoscopic and atomic structures that correspond to a target material property. This framework offers the choice to design scalar-property from the spectroscopic data to steer sample exploration. Our findings reveal correlations of the electronic properties unique to surface terminations, local defect density, and point defects.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2024

Learning Only on Boundaries: A Physics-Informed Neural Operator for Solving Parametric Partial Differential Equations in Complex Geometries

Recently, deep learning surrogates and neural operators have shown promise in solving partial differential equations (PDEs). However, they often require a large amount of training data and are limited to bounded domains. In this work, we present a novel physics-informed neural operator method to solve parameterized boundary value problems without labeled data. By reformulating the PDEs into boundary integral equations (BIEs), we can train the operator network solely on the boundary of the domain. This approach reduces the number of required sample points from $O(N^d)$ to $O(N^{d-1}$), where $d$ is the domain’s dimension, leading to a significant acceleration of the training process. Additionally, our method can handle unbounded problems, which are unattainable for existing physics-informed neural networks (PINNs) and neural operators. Finally, our numerical experiments show the effectiveness of parameterized complex geometries and unbounded problems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Event Classifications on DNE2 Main Experiment Data using a Convolutional Neural Network Ensemble

The Dynamic Networks (DN) Experiment for FY24 (DNE2) is an experiment within DN with the goal of quantitatively evaluating the effectiveness of solutions developed so far by various researchers under the Low Yield Nuclear Monitoring (LYNM) program using a shared set of metrics and datasets. A key component of this experiment is the mimicking of a signature processing pipeline, and comparing currently accepted and standard-use processing methods to more state-of-the-art processes developed under DN. In this work, we focus specifically on the Event Characterization (EC) Focus Area (FA) of the pipeline, where a seismic event’s magnitude, yield and class are identified. We use Deep Learning (DL) to classify the type of events being processed as either earthquakes (EQs) or explosions (EXs) for three iterations of experiment datasets. The model is noticeably more confident and accurate in classifying explosions than earthquakes, reflecting a known shortcoming of the model, that being of a bias towards predicting explosions over earthquakes in the west coast due to training data biases.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Dual-Biometric Human Identification Using Radar Deep Transfer Learning

Accurate human identification using radar has a variety of potential applications, such as surveillance, access control and security checkpoints. Nevertheless, radar-based human identification has been limited to a few motion-based biometrics that are solely reliant on micro-Doppler signatures. This paper proposes for the first time the use of combined radar-based heart sound and gait signals as biometrics for human identification. The proposed methodology starts by converting the extracted biometric signatures collected from 18 subjects to images, and then an image augmentation technique is applied and the deep transfer learning is used to classify each subject. A validation accuracy of 58.7% and 96% is reported for the heart sound and gait biometrics, respectively. Next, the identification results of the two biometrics are combined using the joint probability mass function (PMF) method to report a 98% identification accuracy. To the best of our knowledge, this is the highest reported in the literature to date. Lastly, the trained networks are tested in an actual scenario while being used in an office access control platform to identify different human subjects. We report an accuracy of 76.25%.

47 OTHER INSTRUMENTATION↗
NASA NTRS2016

Decision Analysis Methods Used to Make Appropriate Investments in Human Exploration Capabilities and Technologies

NASA is transforming human spaceflight. The Agency is shifting from an exploration-based program with human activities in low Earth orbit (LEO) and targeted robotic missions in deep space to a more sustainable and integrated pioneering approach. Through pioneering, NASA seeks to address national goals to develop the capacity for people to work, learn, operate, live, and thrive safely beyond Earth for extended periods of time. However, pioneering space involves daunting technical challenges of transportation, maintaining health, and enabling crew productivity for long durations in remote, hostile, and alien environments. Prudent investments in capability and technology developments, based on mission need, are critical for enabling a campaign of human exploration missions. There are a wide variety of capabilities and technologies that could enable these missions, so it is a major challenge for NASA's Human Exploration and Operations Mission Directorate (HEOMD) to make knowledgeable portfolio decisions. It is critical for this pioneering initiative that these investment decisions are informed with a prioritization process that is robust and defensible. It is NASA's role to invest in targeted technologies and capabilities that would enable exploration missions even though specific requirements have not been identified. To inform these investments decisions, NASA's HEOMD has supported a variety of analysis activities that prioritize capabilities and technologies. These activities are often based on input from subject matter experts within the NASA community who understand the technical challenges of enabling human exploration missions. This paper will review a variety of processes and methods that NASA has used to prioritize and rank capabilities and technologies applicable to human space exploration. The paper will show the similarities in the various processes and showcase instances were customer specified priorities force modifications to the process. Specifically, this paper will describe the processes that the NASA Langley Research Center (LaRC) Technology Assessment and Integration Team (TAIT) has used for several years and how those processes have been customized to meet customer needs while staying robust and defensible. This paper will show how HEOMD uses these analyses results to assist with making informed portfolio investment decisions. The paper will also highlight which human exploration capabilities and technologies typically rank high regardless of the specific design reference mission. The paper will conclude by describing future capability and technology ranking activities that will continue o leverage subject matter experts (SME) input while also incorporating more model-based analysis.

Williams-Byrd, Julie↗
DOE OSTI2022

Large Scale Caching and Streaming of Training Data for Online Deep Learning

The training of deep neural network models on large data remains a difficult problem, despite progress towards scalable techniques. In particular, there is a mismatch between the random but predetermined order in which AI flows select training samples and the streaming I/O patterns for which traditional HPC data storage (e.g., parallel file systems) are designed. In addition, as more data are obtained, it is feasible neither simply to train learning models incrementally, due to catastrophic forgetting (i.e., bias towards new samples), nor to train frequently from scratch, due to prohibitive time and/or resource constraints. In this paper, we study data management techniques that combine caching and streaming with rehearsal support in order to enable efficient access to training samples in both offline training and continual learning. We revisit state-of-art streaming approaches based on data pipelines that transparently handle prefetching, caching, shuffling, and data augmentation, and discuss the challenges and opportunities that arise when combining these methods with data-parallel training techniques. We also report on preliminary experiments that evaluate the I/O overheads involved in accessing the training samples from a parallel file system (PFS) under several concurrency scenarios, highlighting the impact of the PFS on the design of the data pipelines.

data pipelines↗
NASA NTRS2018

Virtual Sensors Determined Through Machine Learning

We propose a method that increases the capability of a conventional sensor/instrument, transforming it into an enhanced virtual sensor/instrument. This paper focuses on a virtual thermal IR sensor based on a conventional visual (RGB) sensor. The estimation of thermal IR images can enhance the ability of terrain classification, which is crucial for autonomous navigation of rovers. The estimate in IR from visual band has inherent limitations, as these are different bands, yet correlations between visual/RGB and thermal IR images exist, as different terrains, which visually may appear different, also have different thermal inertia. We developed a deep learning-based algorithm that estimates thermal IR images from RGB images of terrains, providing the feasibility of the idea with average 1.21 error [degree].

Kurazume, Ryo↗
DOE OSTI2025

Deep image reconstruction for background subtraction in heavy-ion collisions

Jet reconstruction in an ultrarelativistic heavy-ion collision suffers from a notoriously large, fluctuating thermal background. Traditional background‐subtraction methods struggle to remove this soft background while preserving the jet's hard substructure. In this Letter, we present DeepSub, a machine-learning-based approach for full-event background‐subtraction. DeepSub utilizes a model based on Swin Transformer layers to denoise jet images and disentangle hard jets from the heavy-ion background. DeepSub significantly outperforms existing subtraction techniques by reproducing key jet observables such as jet 𝑝 T and mass, and substructure observables such as girth and the energy correlation function, at the subpercent to percent level. As such, DeepSub paves the way for precision heavy-ion measurements in hitherto inaccessible kinematic regimes.

Qureshi, Umar Sohail [Stanford University, CA (Uni↗
DOE OSTI2024

Chapter 4: Physically informed deep learning networks for simulating microstructure evolution of 3D polycrystals

As discussed in the previous chapter, high energy diffraction microscopy (HEDM) is used to study the micromechanical evolution of a material during in situ loading. HEDM experiments have been used to verify crystal plasticity (CP) simulations [119, 91, 90, 120], for experimental planning, material design, and to further analyze experimental results. However, Fast Fourier transform-based CP (CP-FFT) or finite element-based CP (CP-FE) methods are often too slow to be used in real-time during an experiment. CP-FFT is faster than CP-FE simulations due to the absence of meshing, but can still take hours to simulate the response of a single volume depending on the size and number of strain steps [127]. Reducing computation time would create a larger exploration space in planning and design, and enable faster analysis of experimental results and real-time feedback during an experiment. This research expands upon previous works to develop a workflow for predicting the full-field evolution of a 3D polycrystal. The workflow is simplified from previous works to predict only orientation and elastic strain tensors (from which stress tensors are calculated). The network is physically informed through loss functions and network architecture for a more robust model. The orientation predictions are informed about the cubic crystal symmetry of the material by incorporating disorientation and misorientation information into the network architecture and loss. The Von Mises stress is used to enforce the correct stress-strain trends in the strain tensor predictions. Additional total strain steps from the elastic and elastoplastic region are included to better capture the stress-strain evolution at smaller total strain steps. Material and hardening parameters are additional inputs into the networks to further inform the network and to study the network’s ability to predict different materials other than those used for training.

36 MATERIALS SCIENCE↗
DOE OSTI2020

DelPlasmid: Finding Plasmids with Deep Learning and Machine Learning v1

Extrachromosomal genetic elements or plasmids are an important driver of DNA exchange and genetic innovation in prokaryotes. The success of plasmids has been attributed to their independent replication from host cell chromosomes and their frequent self-transfer. In recent years, the Department of Energy has recognized the importance of plasmid DNA to the viability and success of microbial communities and JGI has responded with funding to support the research and development of new methods and tools for the efficient sequencing of plasmid DNA. LBNL #2019-037 is responsive to this call.

Andreopoulos, William↗
DOE OSTI2022

Artificial intelligence inferred microstructural properties from voltage–capacity curves

Abstract The quantification of microstructural properties to optimize battery design and performance, to maintain product quality, or to track the degradation of LIBs remains expensive and slow when performed through currently used characterization approaches. In this paper, a convolution neural network-based deep learning approach (CNN) is reported to infer electrode microstructural properties from the inexpensive, easy to measure cell voltage versus capacity data. The developed framework combines two CNN models to balance the bias and variance of the overall predictions. As an example application, the method was demonstrated against porous electrode theory-generated voltage versus capacity plots. For the graphite|LiMn $$_2$$ 2 O $$_4$$ 4 chemistry, each voltage curve was parameterized as a function of the cathode microstructure tortuosity and area density, delivering CNN predictions of Bruggeman’s exponent and shape factor with 0.97 $$R^2$$ R 2 score within 2 s each, enabling to distinguish between different types of particle morphologies, anisotropies, and particle alignments. The developed neural network model can readily accelerate the processing-properties-performance and degradation characteristics of the existing and emerging LIB chemistries.

25 ENERGY STORAGE↗
DOE OSTI2024

RhizoNet segments plant roots to assess biomass and growth for enabling self-driving labs

Abstract Flatbed scanners are commonly used for root analysis, but typical manual segmentation methods are time-consuming and prone to errors, especially in large-scale, multi-plant studies. Furthermore, the complex nature of root structures combined with noisy backgrounds in images complicates automated analysis. Addressing these challenges, this article introduces RhizoNet, a deep learning-based workflow to semantically segment plant root scans. Utilizing a sophisticated Residual U-Net architecture, RhizoNet enhances prediction accuracy and employs a convex hull operation for delineation of the primary root component. Its main objective is to accurately segment root biomass and monitor its growth over time. RhizoNet processes color scans of plants grown in a hydroponic system known as EcoFAB, subjected to specific nutritional treatments. The root detection model using RhizoNet demonstrates strong generalization in the validation tests of all experiments despite variable treatments. The main contributions are the standardization of root segmentation and phenotyping, systematic and accelerated analysis of thousands of images, significantly aiding in the precise assessment of root growth dynamics under varying plant conditions, and offering a path toward self-driving labs.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Predicting Nugget Size of Resistance Spot Welds Using Infrared Thermal Videos With Image Segmentation and Convolutional Neural Network

Resistance spot welding (RSW) is a widely adopted joining technique in automotive industry. Recent advancement in sensing technology makes it possible to collect thermal videos of the weld nugget during RSW using an infrared (IR) camera. The effective and timely analysis of such thermal videos has the potential of enabling in situ nondestructive evaluation (NDE) of the weld nugget by predicting nugget thickness and diameter. Deep learning (DL) has demonstrated to be effective in analyzing imaging data in many applications. However, the thermal videos in RSW present unique data-level challenges that compromise the effectiveness of most pre-trained DL models. We propose a novel image segmentation method for handling the RSW thermal videos to improve the prediction performance of DL models in RSW. The proposed method transforms raw thermal videos into spatial-temporal instances in four steps: video-wise normalization, removal of uninformative images, watershed segmentation, and spatial-temporal instance construction. The extracted spatial-temporal instances serve as the input data for training a DL-based NDE model. The proposed method is able to extract high-quality data with spatial-temporal correlations in the thermal videos, while being robust to the impact of unknown surface emissivity. Overall, our case studies demonstrate that the proposed method achieves better prediction of nugget thickness and diameter than predicting without the transformation.

42 ENGINEERING↗
DOE OSTI2020

Towards Query-Efficient Black-Box Adversary with Zeroth-Order Natural Gradient Descent

Despite the great achievements of the modern deep neural networks (DNNs), the vulnerability/robustness of state-of-the-art DNNs raises security concerns in many application domains requiring high reliability. Various adversarial attacks are proposed to sabotage the learning performance of DNN models. Among those, the black-box adversarial attack methods have received special attentions owing to their practicality and simplicity. Black-box attacks usually prefer less queries in order to maintain stealthy and low costs. However, most of the current black-box attack methods adopt the first-order gradient descent method, which may come with certain deficiencies such as relatively slow convergence and high sensitivity to hyper-parameter settings. In this paper, we propose a zeroth-order natural gradient descent (ZO-NGD) method to design the adversarial attacks, which incorporates the zeroth-order gradient estimation technique catering to the black-box attack scenario and the second-order natural gradient descent to achieve higher query efficiency. The empirical evaluations on image classification datasets demonstrate that ZO-NGD can obtain significantly lower model query complexities compared with state-of-the-art attack methods.

Zhao, Pu↗
DOE OSTI2020

PersGNN: Applying Topological Data Analysis and Geometric Deep Learning to Structure-Based Protein Function Prediction

Understanding protein structure-function relationships is a key challenge in computational biology, with applications across the biotechnology and pharmaceutical industries. While it is known that protein structure directly impacts protein function, many functional prediction tasks use only protein sequence. In this work, we isolate protein structure to make functional annotations for proteins in the Protein Data Bank in order to study the expressiveness of different structure-based prediction schemes. We present PersGNN - an end-to-end trainable deep learning model that combines graph representation learning with topological data analysis to capture a complex set of both local and global structural features. While variations of these techniques have been successfully applied to proteins before, we demonstrate that our hybridized approach, PersGNN, outperforms either method on its own as well as a baseline neural network that learns from the same information. PersGNN achieves a 9.3% boost in area under the precision recall curve (AUPR) compared to the best individual model, as well as high F1 scores across different gene ontology categories, indicating the transferability of this approach.

Swenson, Nicolas↗
DOE OSTI2021

Deep Learning Analysis of Polaritonic Wave Images

Deep learning (DL) is an emerging analysis tool across the sciences and engineering. Encouraged by the successes of DL in revealing quantitative trends in massive imaging data, we applied this approach to nanoscale deeply subdiffractional images of propagating polaritonic waves in complex materials. Utilizing the convolutional neural network (CNN), we developed a practical protocol for the rapid regression of images that quantifies the wavelength and the quality factor of polaritonic waves. Using simulated near-field images as training data, the CNN can be made to simultaneously extract polaritonic characteristics and material parameters in a time scale that is at least 3 orders of magnitude faster than common fitting/processing procedures. The CNN-based analysis was validated by examining the experimental near-field images of charge-transfer plasmon polaritons at graphene/α-RuCl3 interfaces. Our work provides a general framework for extracting quantitative information from images generated with a variety of scanning probe methods.

97 MATHEMATICS AND COMPUTING↗