Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “training 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 1,009 records · Page 56
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

An integrated model-driven method for in-treatment upper airway motion tracking using cine MRI in head and neck radiation therapy

Purpose: For the first time, MRI-guided radiation therapy systems can acquire cine images to dynamically monitor in-treatment internal organ motion. However, the complex head and neck (H&N) structures and low-contrast/resolution of on-board cine MRI images make automatic motion tracking a very challenging task. In this study, the authors proposed an integrated model-driven method to automatically track the in-treatment motion of the H&N upper airway, a complex and highly deformable region wherein internal motion often occurs in an either voluntary or involuntary manner, from cine MRI images for the analysis of H&N motion patterns. Methods: Considering the complex H&N structures and ensuring automatic and robust upper airway motion tracking, the authors firstly built a set of linked statistical shapes (including face, face-jaw, and face-jaw-palate) using principal component analysis from clinically approved contours delineated on a set of training data. The linked statistical shapes integrate explicit landmarks and implicit shape representation. Then, a hierarchical model-fitting algorithm was developed to align the linked shapes on the first image frame of a to-be-tracked cine sequence and to localize the upper airway region. Finally, a multifeature level set contour propagation scheme was performed to identify the upper airway shape change, frame-by-frame, on the entire image sequence. The multifeature fitting energy, including the information of intensity variations, edge saliency, curve geometry, and temporal shape continuity, was minimized to capture the details of moving airway boundaries. Sagittal cine MR image sequences acquired from three H&N cancer patients were utilized to demonstrate the performance of the proposed motion tracking method. Results: The tracking accuracy was validated by comparing the results to the average of two manual delineations in 50 randomly selected cine image frames from each patient. The resulting average dice similarity coefficient (93.28% ± 1.46%) and margin error (0.49 ± 0.12 mm) showed good agreement between the automatic and manual results. The comparison with three other deformable model-based segmentation methods illustrated the superior shape tracking performance of the proposed method. Large interpatient variations of swallowing frequency, swallowing duration, and upper airway cross-sectional area were observed from the testing cine image sequences. Conclusions: The proposed motion tracking method can provide accurate upper airway motion tracking results, and enable automatic and quantitative identification and analysis of in-treatment H&N upper airway motion. By integrating explicit and implicit linked-shape representations within a hierarchical model-fitting process, the proposed tracking method can process complex H&N structures and low-contrast/resolution cine MRI images. Future research will focus on the improvement of method reliability, patient motion pattern analysis for providing more information on patient-specific prediction of structure displacements, and motion effects on dosimetry for better H&N motion management in radiation therapy.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

Feasibility of MRI-only treatment planning for proton therapy in brain and prostate cancers: Dose calculation accuracy in substitute CT images

Purpose: Magnetic resonance imaging (MRI) is increasingly used for radiotherapy target delineation, image guidance, and treatment response monitoring. Recent studies have shown that an entire external x-ray radiotherapy treatment planning (RTP) workflow for brain tumor or prostate cancer patients based only on MRI reference images is feasible. This study aims to show that a MRI-only based RTP workflow is also feasible for proton beam therapy plans generated in MRI-based substitute computed tomography (sCT) images of the head and the pelvis. Methods: The sCTs were constructed for ten prostate cancer and ten brain tumor patients primarily by transforming the intensity values of in-phase MR images to Hounsfield units (HUs) with a dual model HU conversion technique to enable heterogeneous tissue representation. HU conversion models for the pelvis were adopted from previous studies, further extended in this study also for head MRI by generating anatomical site-specific conversion models (a new training data set of ten other brain patients). This study also evaluated two other types of simplified sCT: dual bulk density (for bone and water) and homogeneous (water only). For every clinical case, intensity modulated proton therapy (IMPT) plans robustly optimized in standard planning CTs were calculated in sCT for evaluation, and vice versa. Overall dose agreement was evaluated using dose–volume histogram parameters and 3D gamma criteria. Results: In heterogeneous sCTs, the mean absolute errors in HUs were 34 (soft tissues: 13, bones: 92) and 42 (soft tissues: 9, bones: 97) in the head and in the pelvis, respectively. The maximum absolute dose differences relative to CT in the brain tumor clinical target volume (CTV) were 1.4% for heterogeneous sCT, 1.8% for dual bulk sCT, and 8.9% for homogenous sCT. The corresponding maximum differences in the prostate CTV were 0.6%, 1.2%, and 3.6%, respectively. The percentages of dose points in the head and pelvis passing 1% and 1 mm gamma index criteria were over 91%, 85%, and 38% with heterogeneous, dual bulk, and homogeneous sCTs, respectively. There were no significant changes to gamma index pass rates for IMPT plans first optimized in CT and then calculated in heterogeneous sCT versus IMPT plans first optimized in heterogeneous sCT and then calculated on standard CT. Conclusions: This study demonstrates that proton therapy dose calculations on heterogeneous sCTs are in good agreement with plans generated with standard planning CT. An MRI-only based RTP workflow is feasible in IMPT for brain tumors and prostate cancers.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2019

Accurate and transferable multitask prediction of chemical properties with an atoms-in-molecules neural network

Atomic and molecular properties could be evaluated from the fundamental Schrodinger’s equation and therefore represent different modalities of the same quantum phenomena. Here, we present AIMNet, a modular and chemically inspired deep neural network potential. We used AIMNet with multitarget training to learn multiple modalities of the state of the atom in a molecular system. The resulting model shows on several benchmark datasets state-of-the-art accuracy, comparable to the results of orders of magnitude more expensive DFT methods. It can simultaneously predict several atomic and molecular properties without an increase in the computational cost. With AIMNet, we show a new dimension of transferability: the ability to learn new targets using multimodal information from previous training. The model can learn implicit solvation energy (SMD method) using only a fraction of the original training data and an archive median absolute deviation error of 1.1 kcal/mol compared to experimental solvation free energies in the MNSol database.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Learning the solution operator of parametric partial differential equations with physics-informed DeepONets

Partial differential equations (PDEs) play a central role in the mathematical analysis and modeling of complex dynamic processes across all corners of science and engineering. Their solution often requires laborious analytical or computational tools, associated with a cost that is markedly amplified when different scenarios need to be investigated, for example, corresponding to different initial or boundary conditions, different inputs, etc. In this work, we introduce physics-informed DeepONets, a deep learning framework for learning the solution operator of arbitrary PDEs, even in the absence of any paired input-output training data. We illustrate the effectiveness of the proposed framework in rapidly predicting the solution of various types of parametric PDEs up to three orders of magnitude faster compared to conventional PDE solvers, setting a previously unexplored paradigm for modeling and simulation of nonlinear and nonequilibrium processes in science and engineering.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

When Do Extended Physics-Informed Neural Networks (XPINNs) Improve Generalization?

Physics-informed neural networks (PINNs) have become a popular choice for solving high-dimensional partial differential equations (PDEs) due to their excellent approximation power and generalization ability. Recently, extended PINNs (XPINNs) based on domain decomposition methods have attracted considerable attention due to their effectiveness in modeling multiscale and multiphysics problems and their parallelization. However, theoretical understanding of their convergence and generalization properties remains unexplored. In this study, we take an initial step towards understanding how and when XPINNs outperform PINNs. Specifically, for general multilayer PINNs and XPINNs, we first provide a prior generalization bound via the complexity of the target functions in the PDE problem and a posterior generalization bound via the posterior matrix norms of the networks after optimization. Moreover, based on our bounds, we analyze the conditions under which XPINNs improve generalization. Concretely, our theory shows that the key building block of XPINN, namely, the domain decomposition, introduces a tradeoff for generalization. On the one hand, XPINNs decompose the complex PDE solution into several simple parts, which decreases the complexity needed to learn each part and boosts generalization. On the other hand, decomposition leads to less training data being available in each subdomain, and hence such a model is typically prone to overfitting and may become less generalizable. Empirically, we choose five PDEs to show when XPINNs perform better than, similar to, or worse than PINNs, hence demonstrating and justifying our new theory.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Mixture density network estimation of continuous variable maximum likelihood using discrete training samples

Abstract Mixture density networks (MDNs) can be used to generate posterior density functions of model parameters $$\varvec{\theta }$$ θ given a set of observables $${\mathbf {x}}$$ x . In some applications, training data are available only for discrete values of a continuous parameter $$\varvec{\theta }$$ θ . In such situations, a number of performance-limiting issues arise which can result in biased estimates. We demonstrate the usage of MDNs for parameter estimation, discuss the origins of the biases, and propose a corrective method for each issue.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

Learning-based framework for sensor fault-tolerant building HVAC control with model-assisted learning

As people spend up to 87% of their time indoors, intelligent Heating, Ventilation, and Air Conditioning (HVAC) systems in buildings are essential for maintaining occupant comfort and reducing energy consumption. These HVAC systems in smart buildings rely 'on real-time sensor readings, which in practice often suffer from various faults and could also be vulnerable to malicious attacks. Such faulty sensor inputs may lead to the violation of indoor environment requirements (e.g., temperature, humidity, etc.) and the increase of energy consumption. While many model-based approaches have been proposed in the literature for building HVAC control, it is costly to develop accurate physical models for ensuring their performance and even more challenging to address the impact of sensor faults. In this work, we present a novel learning-based framework for sensor fault-tolerant HVAC control, which includes three deep learning based components for 1) generating temperature proposals with the consideration of possible sensor faults, 2) selecting one of the proposals based on the assessment of their accuracy, and 3) applying reinforcement learning with the selected temperature proposal. Moreover, to address the challenge of training data insufficiency in building-related tasks, we propose a model-assisted learning method leveraging an abstract model of building physical dynamics. Through extensive experiments, we demonstrate that the proposed fault-tolerant HVAC control framework can significantly reduce building temperature violations under a variety of sensor fault patterns while maintaining energy efficiency.

Xu, Shichao↗
DOE OSTI2024

Tandem Predictions for HPC Jobs

At the core of the predictive analytics applied to High Performance Computing (HPC), the most prominent tasks are the prediction of job runtimes and the prediction of job queue times, both of which have the potential for informing HPC users during their every-day decision making. Accurate runtime predictions can help users better choose so-called wallclock times at job submission, decreasing the odds of their jobs waiting in queues longer than necessary. The accurate and timely queue time predictions offered for the available partitions can inform the favorable selection of partitions for running jobs. This potential is well understood as we see in the abundance of research studies that propose solutions for these tasks, including the work published in the last several years. These tasks are seemingly receptive to the Machine Learning (ML) solutions, considering that there is no shortage of training data where HPC centers over time run millions and millions of jobs. However, we study the existing research literature, as well as look for examples in the toolchains supported on the exemplar HPC facilities, and, surprisingly, do not find any practical solutions that are ready to be adopted. We interpret this as a manifestation of the shortage of UX/UI efforts that support HPC analytics and also as a sign that the research has not come to the consensus on solving these tasks. In this study, we aim to shed new light on the long-running task of job queue time prediction by exploring the utility of runtime predictions in improving prediction accuracy and, actually, predicting these two metrics together, in tandem. In other words, we show how runtime predictions become valuable input in the queue time modeling. We challenge the existing approaches to feature engineering for the queue time prediction and describe promising results we obtained for a large dataset of HPC jobs from a supercomputer at the National Renewable Energy Laboratory.

HPC↗
DOE OSTI2025

Using a Large Language Model as a Building Block to Generate Usable Validation and Verification Suite for OpenMP

In the HPC area, both hardware and software move quickly. Often new hardware is developed and deployed, the corresponding software stack, including compilers and other tools, are under active development while leading edge software developers are working to port and tune their applications, all at the same time. While the software ecosystem is in flux, one of the key challenges for users is obtaining insight into the state of implementation of key features in the programming languages and models their applications are using – whether they have been implemented, and whether the implementation conforms to the specification, especially for newly implemented features (less tested by widespread use). OpenMP is one of the most prominent shared memory programming models used for on-node programming in HPC. With the shift towards accelerators (such as GPUs and FPGAs) and heterogeneous programming OpenMP features are getting more complex. It is natural to ask whether generative AI approaches, and large language models (LLMs) in particular, can help in producing validation and verification test suites to allow users better and faster insights into the availability and correctness of OpenMP features of interest. In this work, we explore the use of ChatGPT-4 to generate a suite of tests for OpenMP features. We have chosen a set of directives and clauses, a total of 78 combinations, which first appeared in OpenMP 3.0 (released in May 2008) but are also relevant for accelerators. We prompted ChatGPT to generate tests in the C and Fortran languages, for both host (CPU) and device (accelerator). On the Summit super-computer using the GNU implementation, we found that, of the 78 generated tests 67 C tests and 43 Fortran tests compiled successfully and fewer than those executed to completion. On further analysis we show that not all generated tests are valid. We document the process, results, and provide detailed analysis regarding the quality of tests generated. With the aim of providing input to a production quality validation and verification suite, we manually implement the corrections required to make the tests valid according to the current OpenMP specification. We quantify this effort as small, medium, or large, and record the lines of code changed to correct the invalid tests. With the corrected tests we validate recent implementations from HPE, AMD, and GNU on the Frontier supercomputer. Our experiment and subsequent analysis show that although LLMs are capable of producing HPC specific codes, they are limited by their understanding of the deeper semantics and restrictions of programming models such as OpenMP. Unsurprisingly more commonly used features have better support, while some OpenMP 3.0 directives such as sections and tasking are not universally supported on accelerators. We demonstrate that successful compilation and execution to completion are inadequate metrics for evaluating generated code and that, at this time, commodity LLMs require expert intervention for code verification. This points to gaps in the training data that is currently available for HPC. We demonstrate that with "small" effort 37% of generated invalid C tests and 63% of generated invalid Fortran tests could be corrected. This improves productivity of test generation as we circumvent writing from scratch and the common programming errors associated with it.

Pophale, Swaroop [ORNL] (ORCID:0000000185446367)↗
DOE OSTI2020

Cluster Expansion Analysis of Atomic Order in Li-Ion Battery Cathode Material LiCo y Ni 1-y O 2

A modified cluster-expansion treatment was developed recently to analyze atomic order in LiCo y Ni 1-y O 2 , a model cathode material for Li-ion batteries. In this treatment, referred to as a “spin-atom” cluster expansion, the occupant of a lattice site is identified by its spin state as well as its atomic species. Further, Effective Cluster Interaction (ECI) coefficients are derived from a large training data set (i.e., the set of atomic arrangements for which DFT calculations are performed) which is filtered by an anomaly detection algorithm to eliminate poorly converged DFT calculations. The cluster expansion incorporates Li-Ni (LN) exchange as well as intralayer Co-Ni (CN) exchange. Monte Carlo simulations based on the cluster expansion were applied to the Ni-rich part of the phase diagram. The simulations predict a miscibility gap between y = 0.05 and y = 0.65.

25 ENERGY STORAGE↗
DOE OSTI2026

Conditional Latent Diffusion for High-Resolution Prediction of Electrochemical Surface Morphology

A conditionally guided generative latent diffusion process that is trained on a set of experimental processing parameters and their associated resulting electron microscope images of the electrodeposition process is able to interpolate between processing parameters in a physically consistent way. Electrodeposition of rhenium with pulse and pulse-reverse waveforms is used as a model system, and the process is adaptable to other electrodeposition, electropolishing, or corrosion processes. The method is able to extrapolate, predicting estimates of material morphologies for experimental setups unseen in the training data. The results are demonstrated with experimental data.

36 MATERIALS SCIENCE↗
DOE OSTI2017

MPCpy v0.1

This software can be used to develop and test Model Predictive Control (MPC) algorithms with simulated buildings or deploy MPC in real buildings. This includes the gathering of exogenous data, training of simplified models, and control optimization.

Blum, David↗
DOE OSTI2019

Multi-Task Learning Software Suite

Software suite with multitask learning models. The pipeline includes functionality to process data, train, evaluate models, and generate reports.

Goncalves, Andre↗
DOE OSTI2019

Bayesian Multitask with Structure Learning

Software including a Bayesian multitask learning model. The pipeline includes functionality to process data, train, evaluate models, and generate reports.

Soper, BradenC.↗
DOE OSTI2022

SkX_NN: Neural Network for skyrmion identification in Lorentz TEM

A neural network for segmenting and processing Lorentz TEM images of skyrmion lattices. This software based on U-Net architecture is capable of identifying skyrmions in out-of-focus Lorentz TEM images based on training data. This isspecifically useful for analyzing large area data from in-situ experiments.

PHATAK, CHARUDATTA↗
DOE OSTI2021

Limbo

SAND2021-15055 O Limbo provides a programmatic interface for accessing computer vision training data. It is organized in the Limbo format. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Shead, Timothy↗
DOE OSTI2024

Vistransformers Explained

The Vistransformers Explained library is a collection of python notebooks that demonstrate the internal mechanics and uses of visual-transformer (ViT) machine learning models. The code implements, with mild modifications, ViT models that have been made publicly available through publication and GitHub code. The value added by this code is in-depth explanations of the mathematics behind the sub-modules of the ViT models, including original figures. Additionally, the library contains the code necessary to implement and train the ViT models. The library does not include example training data for the models; instead, it would rely on users generating their own datasets. The code is based on the PyTorch python library. It does not include any files other than python scripts, modules, or notebooks.

Callis, Skylar↗
DOE OSTI2017

pnnl/safekit

Some of the most pressing machine learning applications such as cyber security and object recognition lack enough ground-truth training data to build a classifier. We solve this problem by using unsupervised deep learning techniques to determine when data were anomalous or deviated from the norm. Traditional methods require either specific context dependent expertise to construct models, or prior examples of events and large amounts of well-labeled data to perform machine learning. Some cutting-edge deep learning approaches have been used to characterize complex events, but these have not been applied to the streaming environment and require a great deal of data. We demonstrated the use of an autoencoder and relevant featurization techniques to both learn a feature space and then identify anomalous portions without the aid of labeled data or domain knowledge. In this way, we demonstrated the detection of anomalous features without domain knowledge or tagged examples, and delivered these features to users.

Central, PNNL Developer↗