Search NASA⌕ Search

SEARCH · Search NASA

Results for “Training Principles”

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 NTRS2000

Resistance exercise countermeasures for space flight: implications of training specificity

While resistance exercise should be a logical choice for prevention of strength loss during unloading, the principle of training specificity cannot be overlooked. Our purpose was to explore training specificity in describing the effect of our constant load exercise countermeasure on isokinetic strength performance. Twelve healthy men (mean +/- SD: 28.0 +/- 5.2 years, 179.4 +/- 3.9 cm, 77.5 +/- 13.6 kg) were randomly assigned to no exercise or resistance exercise (REX) during 14 days of bed rest. REX performed five sets of leg press exercise to volitional fatigue (6-10 repetitions) every other day. Unilateral isokinetic concentric-eccentric knee extension testing performed before and on day 15 prior to reambulation included torque-velocity and power-velocity relationships at four velocities (0.52, 1.75, 2.97, and 4.19 rad s-1), torque-position relationship, and contractile work capacity (10 repetitions at 1.05 rad s-1). Two (group) x 2 (time) ANOVA revealed no group x time interactions; thus, groups were combined. Across velocities, angle-specific torque fell 18% and average power fell 20% (p < 0.05). No velocity x time or mode (concentric/eccentric) x time interactions were noted. Torque x position decreased on average 24% (p < 0.05). Total contractile work dropped 27% (p < 0.05). Results indicate bed rest induces rapid and marked reductions in strength and our constant load resistance training protocol did not prevent isokinetic strength losses. Differences between closed-chain training and open-chain testing may explain the lack of protection.

Randomized Controlled Trial↗
DOE OSTI2020

Application of First-Principles-Based Artificial Neural Network Potentials to Multiscale-Shock Dynamics Simulations on Solid Materials

The use of artificial neural network (ANN) potentials trained with first-principles calculations has emerged as a promising approach for molecular dynamics (MD) simulations encompassing large space and time scales while retaining first-principles accuracy. To date, however, the application of ANN-MD has been limited to near-equilibrium processes. Here we combine first-principles-trained ANN-MD with multiscale shock theory (MSST) to successfully describe far-from-equilibrium shock phenomena. Overall, our ANN-MSST-MD approach describes shock-wave propagation in solids with first-principles accuracy but a 5000 times shorter computing time. Accordingly, ANN-MD-MSST was able to resolve fine, long-time elastic deformation at low shock speed, which was impossible with first-principles MD because of the high computational cost. This work thus lays a foundation of ANN-MD simulation to study a wide range of far-from-equilibrium processes.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Biologically-informed excitatory and inhibitory ratio for robust spiking neural network training

Spiking neural networks drawing inspiration from biological constraints of the brain promise an energy-efficient paradigm for artificial intelligence. However, challenges exist in identifying guiding principles to train these networks in a robust fashion. In addition, training becomes an even more difficult problem when incorporating biological constraints of excitatory and inhibitory connections. In this work, we identify several key factors, such as low initial firing rates and diverse inhibitory spiking patterns, that determine the overall ability to train in the context of spiking networks with various ratios of excitatory to inhibitory neurons. The results indicate networks with biologically-realistic excitatory:inhibitory ratios can reliably train at low activity levels and in noisy environments. Additionally, the Van Rossum distance, a measure of spike train synchrony, provides insight into the importance of inhibitory neurons to increase network robustness to noise. This work supports further biologically-informed large-scale networks and energy efficient hardware implementations.

bio-inspired computing↗
DOE OSTI2019

Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks

Advances in computational science offer a principled pipeline for predictive modeling of cardiovascular flows and aspire to provide a valuable tool for monitoring, diagnostics and surgical planning. Such models can be nowadays deployed on large patient-specific topologies of systemic arterial networks and return detailed predictions on flow patterns, wall shear stresses, and pulse wave propagation. However, their success heavily relies on tedious pre-processing and calibration procedures that typically induce a significant computational cost, thus hampering their clinical applicability. Here, we put forth a machine learning framework that enables the seamless synthesis of non-invasive in-vivo measurement techniques and computational flow dynamics models derived from first physical principles. We illustrate this new paradigm by showing how one-dimensional models of pulsatile flow can be used to constrain the output of deep neural networks such that their predictions satisfy the conservation of mass and momentum principles. Once trained on noisy and scattered clinical data of flow and wall displacement, these networks can return physically consistent predictions for velocity, pressure and wall displacement pulse wave propagation, all without the need to employ conventional simulators. A simple post-processing of these outputs can also provide a relatively cheap and effective way for estimating Windkessel model parameters that are required for the calibration of traditional computational models. Finally, the effectiveness of the proposed techniques is demonstrated through a series of prototype benchmarks, as well as a realistic clinical case involving in-vivo measurements near the aorta/carotid bifurcation of a healthy human subject.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2024

Tackling the curse of dimensionality in fractional and tempered fractional PDEs with physics-informed neural networks

Fractional and tempered fractional partial differential equations (PDEs) are effective models of long-range interactions, anomalous diffusion, and non-local effects. Traditional numerical methods for these problems are mesh-based, thus struggling with the curse of dimensionality (CoD). Physics-informed neural networks (PINNs) offer a promising solution due to their universal approximation, generalization ability, and mesh-free training. In principle, Monte Carlo fractional PINN (MC-fPINN) estimates fractional derivatives using Monte Carlo methods and thus could lift CoD. However, this may cause significant variance and errors, hence affecting convergence; in addition, MC-fPINN is sensitive to hyperparameters. In general, numerical methods and specifically PINNs for tempered fractional PDEs are under-developed. Herein, we extend MC-fPINN to tempered fractional PDEs to address these issues, resulting in the Monte Carlo tempered fractional PINN (MC-tfPINN). To reduce possible high variance and errors from Monte Carlo sampling, we replace the one-dimensional (1D) Monte Carlo with 1D Gaussian quadrature, applicable to both MC-fPINN and MC-tfPINN. We validate our methods on various forward and inverse problems of fractional and tempered fractional PDEs, scaling up to 100,000 dimensions. Our improved MC-fPINN/MC-tfPINN using quadrature consistently outperforms the original versions in accuracy and convergence speed in very high dimensions.

42 ENGINEERING↗
DOE OSTI2023

Fast and accurate simulations of calorimeter showers with normalizing flows

In this study, we introduce caloflow, a fast detector simulation framework based on normalizing flows. For the first time, we demonstrate that normalizing flows can reproduce many-channel calorimeter showers with extremely high fidelity, providing a fresh alternative to computationally expensive geant4 simulations, as well as other state-of-the-art fast simulation frameworks based on generative adversarial networks (GANs) or variational autoencoders (VAEs). In addition to the usual histograms of physical features and images of calorimeter showers, we introduce a new metric for judging the quality of generative modeling: the performance of a classifier trained to differentiate real from generated images. We show that GAN-generated images can be identified by the classifier with nearly 100% accuracy, while images generated from caloflow are better able to fool the classifier. More broadly, normalizing flows offer several advantages compared to other state-of-the-art approaches (GANs and VAEs), including tractable likelihoods, stable and convergent training, and principled model selection. Normalizing flows also provide a bijective mapping between data and the latent space, which could have other applications beyond simulation, for example, to detector unfolding.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Classification of Dissolution Events Using Fusion of Effluents Measurements and Classifiers

Classifiers for dissolution events at a radiochemical processing facility are studied using gamma spectra measurements of effluents collected by a high purity germanium detector located at its off-gas stack. Data sets collected at the Oak Ridge National Laboratory’s Radiochemical Engineering Development Center under a Pu dissolution campaign spanning a three months period are utilized. Features corresponding to the activity levels of 15 radionuclides, including isotopes of iodine, krypton, and xenon, that are indicated by the target decay chains, are computed from the spectra at 1 hour intervals. A conceptualization diagram is developed to reflect the steps from the source to measurement to feature computation that depend on fission products indicated by decay chains, chemical processing, and effluents transport to the off-gas stack. A diverse set of eight classifiers based on different design principles are trained using the ground truth data for this campaign, and the outputs of top three classifiers, namely, classification trees, Ensemble of Trees (EOT), and k-nearest neighbor, are combined using EOT classifier-fuser. Our results show that for 5-fold cross validation, features associated with isotopes of xenon provide the lowest classification error among the different elements across the classifiers; the classification error is furthered improved when all 15 isotope features are used by each classifier, and it is again improved by the fusion of three classifiers. Further reduction in classification error is achieved by using a measurement window of 1-2 days which is identified based on half-life time estimates of the isotopes; it is long enough for the stabilization of feature estimates while being short enough not to be affected by the follow on dissolution events. As a net result of feature and classifier fusion, combined with the incorporation of decay chain and isotope half-life information, this approach achieves 98% detection rate while maintaining a false alarm rate under 2% for this data set.

Rao, Nageswara↗
DOE OSTI2021

Improving the Accuracy of Nearest-Neighbor Classification Using Principled Construction and Stochastic Sampling of Training-Set Centroids

A conceptually simple way to classify images is to directly compare test-set data and training-set data. The accuracy of this approach is limited by the method of comparison used, and by the extent to which the training-set data cover configuration space. Here we show that this coverage can be substantially increased using coarse-graining (replacing groups of images by their centroids) and stochastic sampling (using distinct sets of centroids in combination). We use the MNIST and Fashion-MNIST data sets to show that a principled coarse-graining algorithm can convert training images into fewer image centroids without loss of accuracy of classification of test-set images by nearest-neighbor classification. Distinct batches of centroids can be used in combination as a means of stochastically sampling configuration space, and can classify test-set data more accurately than can the unaltered training set. On the MNIST and Fashion-MNIST data sets this approach converts nearest-neighbor classification from a mid-ranking- to an upper-ranking member of the set of classical machine-learning techniques.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Baryonic Effects in Cosmic Shear Tomography: PCA Parameterization and the Importance of Extreme Baryonic Models

Baryonic effects are amongst the most severe systematics to the tomographic analysis of weak lensing data which is the principal probe in many future generations of cosmological surveys like LSST, Euclid etc.. Modeling or parameterizing these effects is essential in order to extract valuable constraints on cosmological parameters. In a recent paper, Eifler et al. (2015) suggested a reduction technique for baryonic effects by conducting a principal component analysis (PCA) and removing the largest baryonic eigenmodes from the data. In this article, we conducted the investigation further and addressed two critical aspects. Firstly, we performed the analysis by separating the simulations into training and test sets, computing a minimal set of principle components from the training set and examining the fits on the test set. We found that using only four parameters, corresponding to the four largest eigenmodes of the training set, the test sets can be fitted thoroughly with an RMS $\sim 0.0011$. Secondly, we explored the significance of outliers, the most exotic/extreme baryonic scenarios, in this method. We found that excluding the outliers from the training set results in a relatively bad fit and degraded the RMS by nearly a factor of 3. Therefore, for a direct employment of this method to the tomographic analysis of the weak lensing data, the principle components should be derived from a training set that comprises adequately exotic but reasonable models such that the reality is included inside the parameter domain sampled by the training set. The baryonic effects can be parameterized as the coefficients of these principle components and should be marginalized over the cosmological parameter space.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Mitigating Extremist Maritime Threats to Radiological Material Transport Vessels

The 21st century has experienced a rise in the threat of global radicalism and extremist organizations, and there has been oft-reported interest from these groups in obtaining or exploiting radiological materials. The threat of extremists acquiring radiological materials while in transit or sabotaging a transport vessel carrying materials at sea is one that requires increased attention. Transport is already one of the most vulnerable times for any cargo but including the difficulties in securing a large ship on the open ocean makes oceanic transport of radiological materials a space of elevated vulnerability for sabotage, theft, or other attacks. The risks of violent extremist attacks on trade ships and trading routes are serious, as shipping routes sustain the global economy, but an attack on a radiological transport vessel could be a magnitude worse, impacting national and international security. While radicalists have largely remained on land in the past due to minimal maritime expertise or inability to project power out of their immediate vicinity, the world has witnessed past attacks on ships such as the USS Cole and the MV Limburg and it is likely only a matter of time before groups extend further into the sea and engage in more maritime campaigns against material transport vessels. This paper suggests methods for increasing safety and security on oceanic transport voyages through behavioral science principles and insider threat recognition training, as well as a better understanding of adversarial and extremist psychology. By providing transport personnel with the knowledge and support to identify and respond to unique maritime threats by extremists, these principles and training suggestions should advance and increase security on the high seas.

Kinney, Justin↗
NASA NTRS1975

Spacelab simulation using a Lear Jet aircraft: Mission no. 4 (ASSESS program)

The fourth ASSESS Spacelab simulation mission utilizing a Lear Jet aircraft featured trained experiment operators (EOs) in place of the participating scientists, to simulate the role and functions of payload specialists in Spacelab who may conduct experiments developed by other scientists. The experiment was a broadband infrared photometer coupled to a 30-cm, open port, IR telescope. No compromises in equipment design or target selection were made to simplify operator tasks; the science goals of the mission were selected to advance the mainline research program of the principle investigator (PI). Training of the EOs was the responsibility of the PI team and consisted of laboratory sessions, on-site training during experiment integration, and integrated mission training using the aircraft as a high-fidelity simulator. The EO permission experience in these several disciplines proved adequate for normal experiment operations, but marginal for the identification and remedy of equipment malfunctions. During the mission, the PI utilized a TV communication system to assist the EOs to overcome equipment difficulties; both science and operations were successfully implemented.

Reller, J. O., Jr.↗
DOE OSTI2020

On-the-fly active learning of interpretable Bayesian force fields for atomistic rare events

Abstract Machine learned force fields typically require manual construction of training sets consisting of thousands of first principles calculations, which can result in low training efficiency and unpredictable errors when applied to structures not represented in the training set of the model. This severely limits the practical application of these models in systems with dynamics governed by important rare events, such as chemical reactions and diffusion. We present an adaptive Bayesian inference method for automating the training of interpretable, low-dimensional, and multi-element interatomic force fields using structures drawn on the fly from molecular dynamics simulations. Within an active learning framework, the internal uncertainty of a Gaussian process regression model is used to decide whether to accept the model prediction or to perform a first principles calculation to augment the training set of the model. The method is applied to a range of single- and multi-element systems and shown to achieve a favorable balance of accuracy and computational efficiency, while requiring a minimal amount of ab initio training data. We provide a fully open-source implementation of our method, as well as a procedure to map trained models to computationally efficient tabulated force fields.

36 MATERIALS SCIENCE↗
DOE OSTI2026

A variational framework for residual-based adaptivity in neural PDE solvers and operator learning

Residual-based adaptive strategies are widely used in scientific machine learning yet remain largely heuristic. We introduce a variational framework that formalizes these methods through convex transformations of the residual, where different transformations correspond to distinct objective functionals. For instance, exponential weights target uniform error minimization, while linear weights recover quadratic error minimization. This perspective reveals adaptive weighting as a means of selecting sampling distributions that optimize a primal objective, directly linking discretization choices to error metrics. This principled approach yields three key benefits: it enables systematic design of adaptive schemes, reduces discretization error by lowering estimator variance, and enhances learning dynamics by improving gradient signal-to-noise ratio. Extending the framework to operator learning, we demonstrate substantial performance gains across diverse optimizers and architectures. Our results provide a theoretical perspective for residual-based adaptivity and establish a foundation for principled discretization and training.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Developing Open-Source Training Materials for AI/ML and Space Biological Sciences Using NASA Cloud-Based Data

Artificial Intelligence (AI) and Machine Learning (ML) has gained significant traction in the biological and biomedical research fields, in part due to a culture of open data sharing and reuse. AI/ML methodology is well-suited to recognize and predict biological patterns from high-dimensional next-generation sequencing data (e.g. whole genome sequencing, transcriptomic sequencing), as well as from biological or medical imaging data (e.g. microscopy, computed tomography, ultrasound, magnetic resonance imaging, radiography). These methodologies hold particular promise for space biosciences research and automated space health monitoring systems. However, there are key considerations for properly training, validating, and testing a machine learning model in biological research or clinical application. Inexperienced researchers can produce models that perform poorly outside of the training dataset. Open Science principles such as data sharing and open-source code must go hand-in-hand with publicly available, high-quality training curricula in best practices, with modules centered on real-life scientific use cases and data so future AI/ML practitioners gain experience on real problems. Here we present the development of open-source training materials for AI/ML and space biosciences, as part of the NASA Transform to Open Science Training (TOPST) initiative. We develop 4 independent training programs, focused on the following topics: 1) Fundamentals of Machine Learning and Space Biosciences Domain, 2) Open Science, Artificial Intelligence, and Ethical Best Practices for Data Sharing and Analysis, 3) Using AI/ML Classification to Identify Gene Networks Affected By Space Exposure in Mouse Liver, and 4) Using Neural Networks to Find DNA Damage Patterns in Immune Cells after Radiation. All programs leverage cloud-based NASA biological datasets. The curriculum we present will enable worldwide access to training in AI/ML and scientific analysis.

James Casaletto↗
NASA NTRS1981

Teaching/learning principles

The potential remote sensing user community is enormous, and the teaching and training tasks are even larger; however, some underlying principles may be synthesized and applied at all levels from elementary school children to sophisticated and knowledgeable adults. The basic rules applying to each of the six major elements of any training course and the underlying principle involved in each rule are summarized. The six identified major elements are: (1) field sites for problems and practice; (2) lectures and inside study; (3) learning materials and resources (the kit); (4) the field experience; (5) laboratory sessions; and (6) testing and evaluation.

Hankins, D. B.↗
DOE OSTI2016

An analysis of image storage systems for scalable training of deep neural networks

This study presents a principled empirical evaluation of image storage systems for training deep neural networks. We employ the Caffe deep learning framework to train neural network models for three different data sets, MNIST, CIFAR-10, and ImageNet. While training the models, we evaluate five different options to retrieve training image data: (1) PNG-formatted image files on local file system; (2) pushing pixel arrays from image files into a single HDF5 file on local file system; (3) in-memory arrays to hold the pixel arrays in Python and C++; (4) loading the training data into LevelDB, a log-structured merge tree based key-value storage; and (5) loading the training data into LMDB, a B+tree based key-value storage. The experimental results quantitatively highlight the disadvantage of using normal image files on local file systems to train deep neural networks and demonstrate reliable performance with key-value storage based storage systems. When training a model on the ImageNet dataset, the image file option was more than 17 times slower than the key-value storage option. Along with measurements on training time, this study provides in-depth analysis on the cause of performance advantages/disadvantages of each back-end to train deep neural networks. We envision the provided measurements and analysis will shed light on the optimal way to architect systems for training neural networks in a scalable manner.

Lim, Seung-Hwan↗