Engineering Papers⌕ Search

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.

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

At least 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS1993

The probabilistic neural network architecture for high speed classification of remotely sensed imagery

In this paper we discuss a neural network architecture (the Probabilistic Neural Net or the PNN) that, to the best of our knowledge, has not previously been applied to remotely sensed data. The PNN is a supervised non-parametric classification algorithm as opposed to the Gaussian maximum likelihood classifier (GMLC). The PNN works by fitting a Gaussian kernel to each training point. The width of the Gaussian is controlled by a tuning parameter called the window width. If very small widths are used, the method is equivalent to the nearest neighbor method. For large windows, the PNN behaves like the GMLC. The basic implementation of the PNN requires no training time at all. In this respect it is far better than the commonly used backpropagation neural network which can be shown to take O(N6) time for training where N is the dimensionality of the input vector. In addition the PNN can be implemented in a feed forward mode in hardware. The disadvantage of the PNN is that it requires all the training data to be stored. Some solutions to this problem are discussed in the paper. Finally, we discuss the accuracy of the PNN with respect to the GMLC and the backpropagation neural network (BPNN). The PNN is shown to be better than GMLC and not as good as the BPNN with regards to classification accuracy.

Chettri, Samir R.↗
NASA NTRS2010

Optimization of a Multi-Stage ATR System for Small Target Identification

An Automated Target Recognition system (ATR) was developed to locate and target small object in images and videos. The data is preprocessed and sent to a grayscale optical correlator (GOC) filter to identify possible regionsof- interest (ROIs). Next, features are extracted from ROIs based on Principal Component Analysis (PCA) and sent to neural network (NN) to be classified. The features are analyzed by the NN classifier indicating if each ROI contains the desired target or not. The ATR system was found useful in identifying small boats in open sea. However, due to "noisy background," such as weather conditions, background buildings, or water wakes, some false targets are mis-classified. Feedforward backpropagation and Radial Basis neural networks are optimized for generalization of representative features to reduce false-alarm rate. The neural networks are compared for their performance in classification accuracy, classifying time, and training time.

false alarm rate↗
DOE OSTI2024

Simultaneously improving accuracy and computational cost under parametric constraints in materials property prediction tasks

Abstract Modern data mining techniques using machine learning (ML) and deep learning (DL) algorithms have been shown to excel in the regression-based task of materials property prediction using various materials representations. In an attempt to improve the predictive performance of the deep neural network model, researchers have tried to add more layers as well as develop new architectural components to create sophisticated and deep neural network models that can aid in the training process and improve the predictive ability of the final model. However, usually, these modifications require a lot of computational resources, thereby further increasing the already large model training time, which is often not feasible, thereby limiting usage for most researchers. In this paper, we study and propose a deep neural network framework for regression-based problems comprising of fully connected layers that can work with any numerical vector-based materials representations as model input. We present a novel deep regression neural network, iBRNet, with branched skip connections and multiple schedulers, which can reduce the number of parameters used to construct the model, improve the accuracy, and decrease the training time of the predictive model. We perform the model training using composition-based numerical vectors representing the elemental fractions of the respective materials and compare their performance against other traditional ML and several known DL architectures. Using multiple datasets with varying data sizes for training and testing, We show that the proposed iBRNet models outperform the state-of-the-art ML and DL models for all data sizes. We also show that the branched structure and usage of multiple schedulers lead to fewer parameters and faster model training time with better convergence than other neural networks. Scientific contribution: The combination of multiple callback functions in deep neural networks minimizes training time and maximizes accuracy in a controlled computational environment with parametric constraints for the task of materials property prediction.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Enabling real-time adaptation of machine learning models at x-ray Free Electron Laser facilities with high-speed training optimized computational hardware

The emergence of novel computational hardware is enabling a new paradigm for rapid machine learning model training. For the Department of Energy’s major research facilities, this developing technology will enable a highly adaptive approach to experimental sciences. In this manuscript we present the per-epoch and end-to-end training times for an example of a streaming diagnostic that is planned for the upcoming high-repetition rate x-ray Free Electron Laser, the Linac Coherent Light Source-II. We explore the parameter space of batch size and data parallel training across multiple Graphics Processing Units and Reconfigurable Dataflow Units. We show the landscape of training times with a goal of full model retraining in under 15 min. Although a full from scratch retraining of a model may not be required in all cases, we nevertheless present an example of the application of emerging computational hardware for adapting machine learning models to changing environments in real-time, during streaming data acquisition, at the rates expected for the data fire hoses of accelerator-based user facilities.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Object Detection and Recognition with PointPillars in LiDAR Point Clouds – Comparisions

In the field of autonomous systems, neural networks have been leveraged for object detection and recognition in 2-dimensional images captured by cameras. Other types of sensors are available for sensing surroundings, including LiDAR sensors, and corresponding networks have been developed to perform detection and recognition in the point clouds generated by these sensors. The approaches are similar, both perform convolutions, but have distinct characteristics and challenges. In designing and configuring autonomous systems, a variety of LiDAR sensors are available, along with configurable deep neural networks to leverage their data. This work presents a review of the PointPillars network, an evolution of the seminal PointNet, comparing accuracy and training time relative to different LiDAR sensors, network and training parameters, CPU and GPU hardware, and the criticality of the use of reflective intensity as a feature. The value of using reflectivity as a predictive feature is explored and quantified to determine if it makes a significant difference in accuracy of the PointPillars network. Two separate LiDAR sensors are utilized, a 16-plane and a 32-plane, and corresponding accuracies and training times with the PointPillars network are evaluated.

LiDAR, machine learning, neural network, object re↗
NASA NTRS2004

Developing a Simulation-Based Training Program for Non-Traditional Caregivers

Simulation-based training enables learning in controlled environments which nevertheless mimic real-world scenarios. It has proven effective in the training of medical personnel and affords rapid assimilation and integration of necessary skills. Non-traditional caregivers often operate in austere environments, where resource and personnel limitations preclude more standard provision of care by highly trained and fully equipped health care teams. In these settings, training time for the caregivers may be limited, with long gaps between time of the training and use of the skills, a limited grasp of the underlying physiology, and unfamiliarity with "medical English" which can render it difficult to communicate concepts to more advanced practitioners when such interaction can take place, as for example, when telemedicine can be used to project medical skills further forward . Methods: Simulation-based training can assist in the familiarization of caregivers to the environment, ensure adequate execution of skills at the appropriate time(s), and allow practice of telemedicine communication patterns between the mentor and caregiver. Results: Scenario-based training can and has been used for initial and sustainment training modules, including self-taught modules for use in the field. Strict identification of the critical concepts is vital, as is development and practice of technically simple procedures wherever possible. Medical devices can off-load tasks from caregivers, as well as to minimize the necessary level of caregiver knowledge, while integrated simulations among all members of the mission team can improve communication and efficiency. Discuss ion: Nontraditional caregivers face unique challenges when learning to provide medical care. Scenario-based curricula allow lesson plans to be tailored to each group's individual needs, as well as being suited for the participation of numerous groups, including the caregiver, evacuation/transport staff, decision-makers, and hospital-based physician.

Bacal, Kira↗
NASA NTRS2012

Rapid Analysis, Self-Calibrating Array for Air Monitoring

Human space missions have critical needs for monitoring and control for life support systems. These systems have monitoring needs that include feedback for closed loop processes and quality control for environmental factors. Sensors and monitoring technologies assure that the air environment and water supply for the astronaut crew habitat fall within acceptable limits, and that the life support system is functioning properly and efficiently. The longer the flight duration and the more distant the destination, the more critical it becomes to have carefully monitored and automated control systems for life support. Past experiments with the JPL ENose have demonstrated a lifetime of the sensor array, with the software, of around 18 months. The lifetime of the calibration, for some analytes, was as long as 24 months. We are working on a sensor array and new algorithms that will include sensor response time in the analysis. The preliminary array analysis for two analytes shows that the analysis time, of an event, can be dropped from 45 minutes to less than10 minutes and array training time can be cut substantially. We will describe the lifetime testing of an array and show lifetime data on individual sensors. This progress will lead to more rapid identification of analytes, and faster training time of the array.

Self Calibrating↗
DOE OSTI2025

Accelerated Over-The-Air Neural Receiver Training Using Self-Contrastive Learning

Self-contrastive learning (SCL), a self-supervised learning method, has been shown to improve image and signal classifier accuracies and reduce the training time for neural communications receivers. In particular, prior work has shown that SCL applied as a pre-training step can improve simulated performance of OFDM in 3GPP TDL channel models by reducing the training time of the downstream classification task (demodulation and demapping). In this work a practical implementation demonstrating SCL pre-training using software defined radios (SDRs) is proposed.

Cooke, Corey [ORNL] (ORCID:0000000234263672)↗
DOE OSTI2026

FitCache: A Transparent Drop-In Framework for Multi-Tier Caching to Accelerate Distributed Deep Learning Workloads

Training in Deep learning (DL) remains highly compute- and data-intensive, with I/O becoming a critical bottleneck as models and datasets scale. Recent studies report that data loading can dominate training time, especially on large-scale HPC systems with shared parallel file systems (PFS). Existing caching approaches either rely on single-tier designs or require intrusive modifications to training pipelines, limiting their portability and effectiveness. In this work, we present FitCache, a transparent drop-in framework for multi-tier caching to accelerate distributed DL training by coordinating fast local memory (e.g., DRAM, Persistent Memory (PMem)) and NVMe as hierarchical caches atop PFS. Our design adapts to hardware diversity, i.e., if NVMe is missing, memory transparently acts as a caching tier, ensuring stable performance. FitCache transparently intercepts I/O requests and issues concurrent fetches across all tiers, returning data from the fastest responder without centralized metadata or static redirection paths. FitCache adapts to dynamic workloads and heterogeneous clusters while maintaining POSIX compatibility. Experiments on Frontier (2048 GPUs) and smaller research clusters show that FitCache reduces training time by up to 40% and per-batch I/O latency by up to 71.6% compared to Lustre Orion PFS, offering a drop-in solution for scalable DL training.

Hu, Guangxing [ORNL] (ORCID:0009000283203614)↗
DOE OSTI2023

Tensor train continuous time solver for quantum impurity models

The simulation of strongly correlated quantum impurity models is a significant challenge in modern condensed matter physics that has multiple important applications. Thus far, the most successful methods for approaching this challenge involve Monte Carlo techniques that accurately and reliably sample perturbative expansions to any order. However, the cost of obtaining high precision through these methods is high. Recently, tensor train decomposition techniques have been developed as an alternative to Monte Carlo integration. In this study, we apply these techniques to the single-impurity Anderson model at equilibrium by calculating the systematic expansion in power of the hybridization of the impurity with the bath. Furthermore, we demonstrate the performance of the method in a paradigmatic application, examining the first-order phase transition on the infinite-dimensional Bethe lattice, which can be mapped to an impurity model through dynamical mean field theory. Our results indicate that using tensor train decomposition schemes allows the calculation of finite-temperature Green's functions and thermodynamic observables with unprecedented accuracy. The methodology holds promise for future applications to frustrated multiorbital systems, using a combination of partially summed series with other techniques pioneered in diagrammatic and continuous time quantum Monte Carlo.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
NASA NTRS2010

Ultrasound Fracture Diagnosis in Space

Introduction: This ground-based investigation accumulated high-level clinical evidence on the sensitivity and specificity of point of care ultrasound performed by expert and novice users for the rapid diagnosis of musculoskeletal (MSK) injuries. We developed preliminary educational methodologies to provide just-in-time training of novice users by creating multi-media training tools and imaging procedures for non expert operators and evaluated the sensitivity and specificity of non-expert performed musculoskeletal ultrasound to diagnose acute injuries in a Level 1 Trauma Center. Methods: Patients with potential MSK injuries were identified in the emergency room. A focused MSK ultrasound was performed by expert operators and compared to standard radiographs. A repeat examination was performed by non-expert operators who received a short, just-in-time multimedia education aid. The sensitivity and specificity of the expert and novice ultrasound examinations were compared to gold standard radiography. Results: Over 800 patients were enrolled in this study. The sensitivity and specificity of expert performed ultrasound exceeded 98% for MSK injuries. Novice operators achieved 97% sensitivity and 99% specificity for targeted examinations with the greatest error in fractures involving the hand and foot. Conclusion: Point of care ultrasound is a sensitive and specific diagnostic test for MSK injury when performed by experts and just-in-time trained novice operators.

Dulchavsky, Scott A.↗
DOE OSTI2023

Efficient learning of power grid voltage control strategies via model-based deep reinforcement learning

Here this article proposes a model-based deep reinforcement learning (DRL) method to design emergency control strategies for short-term voltage stability problems in power systems. Recent advances show promising results for model-free DRL-based methods in power systems control problems. But in power systems applications, these model-free methods have certain issues related to training time (clock time) and sample efficiency; both are critical for making state-of-the-art DRL algorithms practically applicable. DRL-agent learns an optimal policy via a trial-and-error method while interacting with the real-world environment. It is also desirable to minimize the direct interaction of the DRL agent with the real-world power grid due to its safety-critical nature. Additionally, the state-of-the-art DRL-based policies are mostly trained using a physics-based grid simulator where dynamic simulation is computationally intensive, lowering the training efficiency. We propose a novel model-based DRL framework where a deep neural network (DNN)-based dynamic surrogate model (SM), instead of a real-world power grid or physics-based simulation, is utilized within the policy learning framework, making the process faster and more sample efficient. However, having stable training in model-based DRL is challenging because of the complex system dynamics of large-scale power systems. We addressed these issues by incorporating imitation learning to have a warm start in policy learning, reward-shaping, and multi-step loss in surrogate model training. Finally, we achieved 97.5% reduction in samples and 87.7% reduction in training time for an application to the IEEE 300-bus test system.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Language models for the prediction of SARS-CoV-2 inhibitors

The COVID-19 pandemic highlights the need for computational tools to automate and accelerate drug design for novel protein targets. We leverage deep learning language models to generate and score drug candidates based on predicted protein binding affinity. We pre-trained a deep learning language model (BERT) on ∼9.6 billion molecules and achieved peak performance of 603 petaflops in mixed precision. Our work reduces pre-training time from days to hours, compared to previous efforts with this architecture, while also increasing the dataset size by nearly an order of magnitude. For scoring, we fine-tuned the language model using an assembled set of thousands of protein targets with binding affinity data and searched for inhibitors of specific protein targets, SARS-CoV-2 Mpro and PLpro. We utilized a genetic algorithm approach for finding optimal candidates using the generation and scoring capabilities of the language model. Our generalizable models accelerate the identification of inhibitors for emerging therapeutic targets.

Blanchard, Andrew E.↗
NASA NTRS1995

Helicopter Human Factors

Even under optimal conditions, helicopter flight is a most demanding form of human-machine interaction, imposing continuous manual, visual, communications, and mental demands on pilots. It is made even more challenging by small margins for error created by the close proximity of terrain in NOE flight and missions flown at night and in low visibility. Although technology advances have satisfied some current and proposed requirements, hardware solutions alone are not sufficient to ensure acceptable system performance and pilot workload. However, human factors data needed to improve the design and use of helicopters lag behind advances in sensor, display, and control technology. Thus, it is difficult for designers to consider human capabilities and limitations when making design decisions. This results in costly accidents, design mistakes, unrealistic mission requirements, excessive training costs, and challenge human adaptability. NASA, in collaboration with DOD, industry, and academia, has initiated a program of research to develop scientific data bases and design principles to improve the pilot/vehicle interface, optimize training time and cost, and maintain pilot workload and system performance at an acceptable level. Work performed at Ames, and by other research laboratories, will be reviewed to summarize the most critical helicopter human factors problems and the results of research that has been performed to: (1) Quantify/model pilots use of visual cues for vehicle control; (2) Improve pilots' performance with helmet displays of thermal imagery and night vision goggles for situation awareness and vehicle control; (3) Model the processes by which pilots encode maps and compare them to the visual scene to develop perceptually and cognitively compatible electronic map formats; (4) Evaluate the use of spatially localized auditory displays for geographical orientation, target localization, radio frequency separation; (5) Develop and flight test control/display concepts; (6) Quantify, model, predict, and improve pilots, workload-management strategies; and (7) Design computer-game trainers to reduce training time and cost.

Hart, Sandra G.↗
NASA NTRS2005

Evaluation of shoulder integrity in space: first report of musculoskeletal US on the International Space Station

Investigative procedures were approved by Henry Ford Human Investigation Committee and NASA Johnson Space Center Committee for Protection of Human Subjects. Informed consent was obtained. Authors evaluated ability of nonphysician crewmember to obtain diagnostic-quality musculoskeletal ultrasonographic (US) data of the shoulder by following a just-in-time training algorithm and using real-time remote guidance aboard the International Space Station (ISS). ISS Expedition-9 crewmembers attended a 2.5-hour didactic and hands-on US training session 4 months before launch. Aboard the ISS, they completed a 1-hour computer-based Onboard Proficiency Enhancement program 7 days before examination. Crewmembers did not receive specific training in shoulder anatomy or shoulder US techniques. Evaluation of astronaut shoulder integrity was done by using a Human Research Facility US system. Crew used special positioning techniques for subject and operator to facilitate US in microgravity environment. Common anatomic reference points aided initial probe placement. Real-time US video of shoulder was transmitted to remote experienced sonologists in Telescience Center at Johnson Space Center. Probe manipulation and equipment adjustments were guided with verbal commands from remote sonologists to astronaut operators to complete rotator cuff evaluation. Comprehensive US of crewmember's shoulder included transverse and longitudinal images of biceps and supraspinatus tendons and articular cartilage surface. Total examination time required to guide astronaut operator to acquire necessary images was approximately 15 minutes. Multiple arm and probe positions were used to acquire dynamic video images that were of excellent quality to allow evaluation of shoulder integrity. Postsession download and analysis of high-fidelity US images collected onboard demonstrated additional anatomic detail that could be used to exclude subtle injury. Musculoskeletal US can be performed in space by minimally trained operators by using remote guidance. This technique can be used to evaluate shoulder integrity in symptomatic crewmembers after strenuous extravehicular activities or to monitor microgravity-associated changes in musculoskeletal anatomy. Just-in-time training, combined with remote experienced physician guidance, may provide a useful approach to complex medical tasks performed by nonexperienced personnel in a variety of remote settings, including current and future space programs. (c) RSNA, 2004.

Space Flight/education↗
NASA NTRS2009

Acute Hemodynamic Effects of the Braslet-M Device on the International Space Station

The Braslet-M occlusion device is prescribed for cosmonauts as a countermeasure for early phases of spaceflight to temporarily alleviate symptoms associated with the cephalad fluid shift. Using a multipurpose ultrasound (US) device onboard, we assessed the acute hemodynamic effects of the Bracelet-M device on a long duration International Space Station (ISS) crewmember. Methods A combination of just-in-time training and real-time remote expert assistance was used to conduct the imaging procedures. An HDI-5000 imager (Philips, Bothell, WA) was used, provided by the ISS Human Research Facility. Superficial femoral artery (SFA), femoral vein (FV) flow spectra were obtained at mid-thigh level. Left ventricle was imaged through the apical 4-chamber view, with Color M-Mode to measure propagation velocity (V (p)). After 10 minutes of Bracelet-M use, data collection was repeated. All data were transmitted in DICOM format to ground for analysis. Results With Braslet-M, cardiac V(p) slope decreased (56ms to 42ms). A stagnation signature in the FV was seen suggesting impeded flow (rouleaux formation, too-low-to-measure velocity, and increase in diameter). Quadri-phasic flow in SFA was seen both before and after Braslet-M application. Velocities in the SFA decreased with Braslet-M (65cm/sec to 52cm/sec) and so did the time velocity integrals (16.97 to 12.4); the flow pattern spoke of resistivity increase in the vascular bed. Conclusion In the long duration ISS crewmember we observed effects of lower extremity venous occlusion through both central and peripheral indicators. A part of circulating volume transferred to peripheral potential vascular space. Impediment to venous outflow was demonstrated objectively, with a commensurate change in the flow pattern of the main feeding artery. Central volume reduction caused lower V(p). Additional studies are warranted to determine the time course of the changes and the dynamics in interstitial fluid sequestration, as well as the safe levels and duration of the compression forces.

Hamilton, Douglas R.↗