SEARCH · Engineering Papers

Results for “Training Time”

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 667 records · Page 37

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Multiscale Data-Driven Seismic Full-Waveform Inversion With Field Data Study

Seismic full-waveform inversion (FWI), which uses iterative methods to estimate high-resolution subsurface models from seismograms, is a powerful imaging technique in exploration geophysics. In recent years, the computational cost of FWI has grown exponentially due to the increasing size and resolution of seismic data. Moreover, it is a nonconvex problem and can encounter local minima due to the limited accuracy of the initial velocity models or the absence of low frequencies in the measurements. To overcome these computational issues, we develop a multiscale data-driven FWI method based on fully convolutional networks (FCNs). In preparing the training data, we first develop a real-time style transform method to create a large set of synthetic subsurface velocity models from natural images. We then develop two convolutional neural networks with encoder-decoder structures to reconstruct the low- and high-frequency components of the subsurface velocity models, separately. To validate the performance of our data-driven inversion method and the effectiveness of the synthesized training set, we compare it with conventional physics-based waveform inversion approaches using both synthetic and field data. Finally, these numerical results demonstrate that, once our model is fully trained, it can significantly reduce the computation time and yield more accurate subsurface velocity models in comparison with conventional FWI.

58 GEOSCIENCES↗

DOE OSTI2020

Hindsight logging for model training

In modern Machine Learning, model training is an iterative, experimental process that can consume enormous computation resources and developer time. To aid in that process, experienced model developers log and visualize program variables during training runs. Exhaustive logging of all variables is infeasible, so developers are left to choose between slowing down training via extensive conservative logging, or letting training run fast via minimalist optimistic logging that may omit key information. As a compromise, optimistic logging can be accompanied by program checkpoints; this allows developers to add log statements post-hoc, and "replay" desired log statements from checkpoint---a process we refer to as hindsight logging. Unfortunately, hindsight logging raises tricky problems in data management and software engineering. Done poorly, hindsight logging can waste resources and generate technical debt embodied in multiple variants of training code. In this paper, we present methodologies for efficient and effective logging practices for model training, with a focus on techniques for hindsight logging. Our goal is for experienced model developers to learn and adopt these practices. To make this easier, we provide an open-source suite of tools for Fast Low-Overhead Recovery (flor) that embodies our design across three tasks: (i) efficient background logging in Python, (ii) adaptive periodic checkpointing, and (iii) an instrumentation library that codifies hindsight logging for efficient and automatic record-replay of model-training. Model developers can use each flor tool separately as they see fit, or they can use flor in hands-free mode, entrusting it to instrument their code end-to-end for efficient record-replay. Our solutions leverage techniques from physiological transaction logs and recovery in database systems. Evaluations on modern ML benchmarks demonstrate that flor can produce fast checkpointing with small user-specifiable overheads (e.g. 7%), and still provide hindsight log replay times orders of magnitude faster than restarting training from scratch.

Computer Science↗

NASA NTRS1987

Rapid prototyping and AI programming environments applied to payload modeling

This effort focused on using artificial intelligence (AI) programming environments and rapid prototyping to aid in both space flight manned and unmanned payload simulation and training. Significant problems addressed are the large amount of development time required to design and implement just one of these payload simulations and the relative inflexibility of the resulting model to accepting future modification. Results of this effort have suggested that both rapid prototyping and AI programming environments can significantly reduce development time and cost when applied to the domain of payload modeling for crew training. The techniques employed are applicable to a variety of domains where models or simulations are required.

Carnahan, Richard S., Jr.↗

NASA NTRS1985

Ambulatory Feedback System

This presentation discusses instrumentation that will be used for a specific event, which we hope will carry on to future events within the Space Shuttle program. The experiment is the Autogenic Feedback Training Experiment (AFTE) scheduled for Spacelab 3, currently scheduled to be launched in November, 1984. The objectives of the AFTE are to determine the effectiveness of autogenic feedback in preventing or reducing space adaptation syndrome (SAS), to monitor and record in-flight data from the crew, to determine if prediction criteria for SAS can be established, and, finally, to develop an ambulatory instrument package to mount the crew throughout the mission. The purpose of the Ambulatory Feedback System (AFS) is to record the responses of the subject during a provocative event in space and provide a real-time feedback display to reinforce the training.

Finger, Herbert↗

DOE OSTI2023

classLog: Logistic regression for the classification of genetic sequences

Introduction Sequencing and phylogenetic classification have become a common task in human and animal diagnostic laboratories. It is routine to sequence pathogens to identify genetic variations of diagnostic significance and to use these data in realtime genomic contact tracing and surveillance. Under this paradigm, unprecedented volumes of data are generated that require rapid analysis to provide meaningful inference. Methods We present a machine learning logistic regression pipeline that can assign classifications to genetic sequence data. The pipeline implements an intuitive and customizable approach to developing a trained prediction model that runs in linear time complexity, generating accurate output rapidly, even with incomplete data. Our approach was benchmarked against porcine respiratory and reproductive syndrome virus (PRRSv) and swine H1 influenza A virus (IAV) datasets. Trained classifiers were tested against sequences and simulated datasets that artificially degraded sequence quality at 0, 10, 20, 30, and 40%. Results When applied to a poor-quality sequence data, the classifier achieved between >85% to 95% accuracy for the PRRSv and the swine H1 IAV HA dataset and this increased to near perfect accuracy when using the full dataset. The model also identifies amino acid positions used to determine genetic clade identity through a feature selection ranking within the model. These positions can be mapped onto a maximum-likelihood phylogenetic tree, allowing for the inference of clade defining mutations. Discussion Our approach is implemented as a python package with code available at https://github.com/flu-crew/classLog .

Zeller, Michael A.↗

DOE OSTI2021

Burst-mode dual-comb spectroscopy

We introduce a new modality of dual-comb spectroscopy that enables a simplified and powerful new approach for time-resolved measurements with increased acquisition rates. This "burst mode" form of dual-comb spectroscopy (BM-DCS) relies on the multiplexing of each probe pulse into a short train of pulses. With this approach we demonstrate a time-resolved series of absorption-based spectroscopic measurements of a laser-induced plasma using only a single laser ablation shot and identify 22 Nd lines not previously reported in the literature. The transmission spectra spanned 3.1 THz and were acquired at an effective acquisition rate of 25 kHz with 40 ms time-resolution. This simple modification to 100 MHz level dual-comb systems provides a flexible approach for studying transient and low duty-cycle events such as laser-induced plasmas, combustion, and explosive reactions.

(280.5395) Plasma Diagnostics, broadband laser spe↗

NASA NTRS1976

Continuation of advanced crew procedures development techniques

An operational computer program, the Procedures and Performance Program (PPP) which operates in conjunction with the Phase I Shuttle Procedures Simulator to provide a procedures recording and crew/vehicle performance monitoring capability was developed. A technical synopsis of each task resulting in the development of the Procedures and Performance Program is provided. Conclusions and recommendations for action leading to the improvements in production of crew procedures development and crew training support are included. The PPP provides real-time CRT displays and post-run hardcopy output of procedures, difference procedures, performance data, parametric analysis data, and training script/training status data. During post-run, the program is designed to support evaluation through the reconstruction of displays to any point in time. A permanent record of the simulation exercise can be obtained via hardcopy output of the display data and via transfer to the Generalized Documentation Processor (GDP). Reference procedures data may be transferred from the GDP to the PPP. Interface is provided with the all digital trajectory program, the Space Vehicle Dynamics Simulator (SVDS) to support initial procedures timeline development.

Arbet, J. D.↗

DOE OSTI2022

Three-dimensional realizations of flood flow in large-scale rivers using the neural fuzzy-based machine-learning algorithms

Machine learning methods have been extensively used to study the dynamics of complex fluid flows. One such algorithm, known as adaptive neural fuzzy inference system (ANFIS), can generate data-driven predictions for flow fields, but has not been applied to natural geophysical flows in large-scale rivers. Herein, we demonstrate the potential of ANFIS to produce three-dimensional (3D) realizations of the instantaneous flood flow field in several large-scale, virtual meandering rivers. The 3D dynamics of flood flow in large-scale rivers were obtained using large-eddy simulation (LES). The LES results, i.e., the 3D velocity components, were employed to train the learnable coefficients of an ANFIS. Further, the trained ANFIS, along with a few time-steps of LES results (precursor data) were then used to produce 3D realizations of flood flow fields in large-scale rivers with geometries other than the one the ANFIS was trained with. We also used the trained ANFIS to generate 3D realizations of river flow at a discharge other than that the ANFIS was trained with. The flow field results obtained from ANFIS were validated using separate LES runs to assess the accuracy of the 3D instantaneous realizations of the machine learning algorithm. An error analysis was conducted to quantify the discrepancies among the ANFIS and LES results for various flood flow predictions in large-scale rivers.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS2001

Emotional/Mental Challenges Pre-, In-, and Post-Flight

Dr. Voss has flown aboard the Space Shuttle five times. She knows well her inner concerns, emotions, and mental challenges attending such highly demanding and risky adventures. And she has shared those ideas with her colleagues. She notes that their busy training schedules and fully committed on orbit time allow little time for dwelling on most of these issues. However, they are nonetheless real and may not be ignored with impunity. She thinks that perhaps they are more striking for rookie space farers, but all spacecrew members share them and can profit by assuring proper support and unique solutions for their own specific situation, which could vary with the mission. In her own experience, she found notable benefit from sharing with close members of her family, both before flight and during. The latter has proved of great value to all crew persons in the form of their personal ground contact time with family and friends. In addition, how one arranges and what one provides in the on board personal space and time goes far toward keeping a confident and upbeat view of the big picture. The type and amount of off duty diversions (e.g., music, reading material) are important, as are how one participates in group time. And it is universally agreed that viewing time at the spacecraft windows offers great joy and calm. Dr. Voss conjectures that there could be a difference in how people deal with these matters on busy, short-duration (Shuttle type) missions versus those of longer ones, particularly out of low earth orbit, where the options in the advent of mishap are fewer. Her final opinion is one of optimism and assurance that the human person will do well in coping with this new environment.

Voss, Janice↗

DOE OSTI2022

Time-lapse seismic data inversion for estimating reservoir parameters using deep learning

Geologic carbon sequestration involves the injection of captured carbon dioxide ([Formula: see text]) into subsurface formations for long-term storage. The movement and fate of the injected [Formula: see text] plume is of great concern to regulators because monitoring helps to identify potential leakage zones and determines the possibility of safe long-term storage. To address this concern, we design a deep-learning framework for [Formula: see text] saturation monitoring to determine the geologic controls on the storage of the injected [Formula: see text]. We use different combinations of porosities and permeabilities for a given reservoir to generate saturation and velocity models. We train the deep-learning model with a few time-lapse seismic images and their corresponding changes in saturation values for a particular [Formula: see text] injection site. The deep-learning model learns the mapping from the change in the time-lapse seismic response to the change in [Formula: see text] saturation during the training phase. We then apply the trained model to data sets comprising different time-lapse seismic image slices (corresponding to different time instances) generated using different porosity and permeability distributions that are not part of the training to estimate the [Formula: see text] saturation values along with the plume extent. Our algorithm provides a deep-learning assisted framework for the direct estimation of [Formula: see text] saturation values and plume migration in heterogeneous formations using the time-lapse seismic data. Our method improves the efficiency of time-lapse inversion by streamlining the large number of intermediate steps in the conventional time-lapse inversion workflow. This method also helps to incorporate the geologic uncertainty for a given reservoir by accounting for the statistical distribution of porosity and permeability during the training phase. Tests on different examples verify the effectiveness of our approach.

Geochemistry & Geophysics↗

DOE OSTI2025

Breaking the barrier of human-annotated training data for machine learning-aided plant research using aerial imagery

Machine learning (ML) can accelerate biological research. However, the adoption of such tools to facilitate phenotyping based on sensor data has been limited by (i) the need for a large amount of human-annotated training data for each context in which the tool is used and (ii) phenotypes varying across contexts defined in terms of genetics and environment. This is a major bottleneck because acquiring training data is generally costly and time-consuming. This study demonstrates how a ML approach can address these challenges by minimizing the amount of human supervision needed for tool building. A case study was performed to compare ML approaches that examine images collected by an uncrewed aerial vehicle to determine the presence/absence of panicles (i.e. “heading”) across thousands of field plots containing genetically diverse breeding populations of 2 Miscanthus species. Automated analysis of aerial imagery enabled the identification of heading approximately 9 times faster than in-field visual inspection by humans. Leveraging an Efficiently Supervised Generative Adversarial Network (ESGAN) learning strategy reduced the requirement for human-annotated data by 1 to 2 orders of magnitude compared to traditional, fully supervised learning approaches. The ESGAN model learned the salient features of the data set by using thousands of unlabeled images to inform the discriminative ability of a classifier so that it required minimal human-labeled training data. This method can accelerate the phenotyping of heading date as a measure of flowering time in Miscanthus across diverse contexts (e.g. in multistate trials) and opens avenues to promote the broad adoption of ML tools.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2025

pixelvar79/ESGAN-Flowering-Detection-paper

Varela, Sebastian↗

DOE OSTI2021

Adaptive machine learning for time-varying systems: low dimensional latent space tuning

Machine learning (ML) tools such as encoder-decoder convolutional neural networks (CNN) can represent incredibly complex nonlinear functions which map between combinations of images and scalars. For example, CNNs can be used to map combinations of accelerator parameters and images which are 2D projections of the 6D phase space distributions of charged particle beams as they are transported between various particle accelerator locations. Despite their strengths, applying ML to time-varying systems, or systems with shifting distributions, is an open problem, especially for large systems for which collecting new data for re-training is impractical or interrupts operations. Particle accelerators are one example of large time-varying systems for which collecting detailed training data requires lengthy dedicated beam measurements which may no longer be available during regular operations. We present a novel method of adaptive ML for time-varying systems. Our approach is to map very high (N ≈ 100k) dimensional inputs (a combination of scalar parameters and images) into the low dimensional (N ≈ 2) latent space at the output of the encoder section of an encoder-decoder CNN. We then actively tune the low dimensional latent space-based representation of complex system dynamics by the addition of an adaptively tuned feedback vector directly before the decoder sections builds back up to our image-based high-dimensional phase space density representations. This method allows us to learn correlations within and to quickly tune the characteristics of incredibly large parameter space systems and to track their evolution in real time based on feedback without massive new data sets for re-training. We demonstrate that our method can accurately predict and track the phase space of charged particle beams at various locations in a particle accelerator by adaptively adjusting in real-time while the unknown input beam distribution of the accelerator is changing in shape, charge, and offset and while the RF system of the accelerator itself is also changing in an unpredictable way. For FACET-II we demonstrate that such an approach has the potential to use transverse deflecting cavity and energy spread spectrum beam measurements to accurately predict 2D projections of the 6D phase space of the electron beam at the plasma wakefield acceleration interaction point where such diagnostics are unavailable.

47 OTHER INSTRUMENTATION↗

DOE OSTI2022

Towards Compact Neural Networks via End-to-End Training: A Bayesian Tensor Approach with Automatic Rank Determination

Post-training model compression can reduce the inference costs of deep neural networks, but uncompressed training still consumes enormous hardware resources and energy. To enable low-energy training on edge devices, it is highly desirable to directly train a compact neural network from scratch with a low memory cost. Low-rank tensor decomposition is an effective approach to reduce the memory and computing costs of large neural networks. However, directly training low-rank tensorized neural networks is a very challenging task because it is hard to determine a proper tensor rank a priori, and the tensor rank controls both model complexity and accuracy. Here, this paper presents a novel end-to-end framework for low-rank tensorized training. We first develop a Bayesian model that supports various low-rank tensor formats (e.g., CANDECOMP/PARAFAC, Tucker, tensor-train, and tensor-train matrix) and reduces neural network parameters with automatic rank determination during training. Then we develop a customized Bayesian solver to train large-scale tensorized neural networks. Our training methods shows orders-of-magnitude parameter reduction and little accuracy loss (or even better accuracy) in the experiments. On a very large deep learning recommendation system with over 4.2 ×10 9 model parameters, our method can reduce the parameter number to 1.6 ×10 5 automatically in the training process (i.e., by 2.6 ×10 4 times) while achieving almost the same accuracy. Code is available at https://github.com/colehawkins/bayesian-tensor-rank-determination.

compact neural networks↗

DOE OSTI2024

Georgia Tech Accelerated, Compressed, and Regularized Compute of Kinetic-based PDEs (Final Report)

This report summarizes the collaborative effort between Lawrence Livermore National Laboratory and Georgia Tech to enhance the BoBa library for tensor train computation in PDE solvers, with a target on kinetic equations and their continuum limits. We aimed to reduce computational cost and memory usage by replacing traditional array-based computations with tensor trains. We examined the compressibility of time-evolving solutions to the Euler equations with discontinuities. We also explored using the first invsicid and linear regularization of the compressible flow equations via the information geometric regularization (IGR). We explored this in a tensor train formulation. To identify that inverse terms in the IGR equations pose problems for tensor train formulations and investigate efficient methods for batched inversion of tensor trains.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

Transferable Reinforcement Learning for Smart Homes: Preprint

To harness the great amount of untapped resources at the demand side, smart home technology plays a vital role in solving the "last mile" problem in smart grid. Reinforcement learning (RL), which has demonstrated an outstanding performance in solving many sequential decision-making problems, can be a great candidate to be used in smart home control. For instance, many studies have started investigating the load scheduling problem under dynamic pricing scheme. Based on those, this study aims at providing an affordable solution to encourage a higher smart home adoption rate. Specifically, we investigate combining transfer learning (TL) with RL to reduce the training cost of an optimal RL control policy. Given an optimal policy for a benchmark home, TL can jump-start the RL training of a policy for a new home, which has different appliances and user preferences. Simulation results show that by leveraging TL, RL training converges faster and requires much less computing time for new homes that are similar to the benchmark home. In all, this study proposes a cost-effective approach for training RL control policies for homes at scale, which ultimately reduces the controller's implementation costs, increases the adoption rate of RL controllers, and makes more homes grid-interactive.

ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION,↗

DOE OSTI2021

Characterizing the acceleration time of laser-driven ion acceleration with data-informed neural networks

Peak ion energy is an important figure-of-merit in short-pulse, laser-driven ion acceleration and is dependent on an associated acceleration time. Standard metrics for these quantities depend on analytical results such as the self-similar fluid model or empirical models based on relatively small experimental and simulation datasets. In this work we attempt to use a data-informed neural network (NN) as a surrogate model for a large ensemble of PIC simulations to investigate an effective acceleration time. We explore the application of a stacked convolutional and recurrent NN architecture for improved regression by incorporating the time dependencies of the data into the training process. Of particular note is how pretraining a network on lower fidelity data, e.g. 1D analytical results, greatly improves the network's ability to learn more complex, higher fidelity data. Finally, the dependency of the acceleration time on various laser and plasma parameters is explored.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

NASA NTRS2004

Results from an Investigation into Extra-Vehicular Activity (EVA) Training Related Shoulder Injuries

The number and complexity of extravehicular activities (EVAs) required for the completion and maintenance of the International Space Station (ISS) is unprecedented. The training required to successfully complete this magnitude of space walks presents a real risk of overuse musculoskeletal injuries to the EVA crew population. There was mounting evidence raised by crewmembers, trainers, and physicians at the Johnson Space Center (JSC) between 1999 and 2002 that suggested a link between training in the Neutral - Buoyancy Lab (NBL) and the several reported cases of shoulder injuries. The short- and long-term health consequences of shoulder injury to astronauts in training as well as the potential mission impact associated with surgical intervention to assigned EVA crew point to this as a critical problem that must be mitigated. Thus, a multi-directorate tiger team was formed in December of 2002 led by the EVA Office and Astronaut Office at the JSC. The primary objectives of this Tiger Team were to evaluate the prevalence of these injuries and substantiate the relationship to training in the NBL with the crew person operating in the EVA Mobility Unit (EMU). Between December 2002 and June of 2003 the team collected data, surveyed crewmembers, consulted with a variety of physicians, and performed tests. The results of this effort were combined with the vast knowledge and experience of the Tiger Team members to formulate several findings and over fifty recommendations. This paper summarizes those findings and recommendations as well as the process by which these were determined. The Tiger Team concluded that training in the NBL was directly linked to several major and minor shoulder injuries that had occurred. With the assistance of JSC flight surgeons, outside consultants, and the lead crewmember/physician on the team, the mechanisms of injury were determined. These mechanisms were then linked to specific aspects of the hardware design, operational techniques, and the training environment. During the 1999 to 2003 time frame many variables converged to make it impossible to determine with any accuracy which one or two root causes were primarily involved. Therefore a broad range of recommendations was established to prevent future injury to crewmembers training in the NBL in the near term. Many of these recommendations are lessons learned that are essentially timeless and therefore should be passed on to future EVA endeavors to ensure that hardware designs and operational techniques utilized in the future consider the demands of training on the human body here on earth.

Johnson, Brian J.↗