Search NASA⌕ Search

SEARCH · Search NASA

Results for “successive training”

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 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Application of machine learning to characterize gas hydrate reservoirs in Mackenzie Delta (Canada) and on the Alaska north slope (USA)

Here, artificial neural network-trained models were used to predict gas hydrate saturation distributions in permafrost-associated deposits in the Eileen Gas Hydrate Trend on the Alaska North Slope (ANS), USA and at the Mallik research site in the Beaufort-Mackenzie Basin, Northwest Territories, Canada. The database of Logging-While-Drilling (LWD) and wireline logs collected at five wells (Mount Elbert, Ignik Sikumi, and Kuparuk 7–11–12 wells at ANS, plus 2L-38 and 5L-38 wells at the Mallik research site) includes more than 10,000 depth points, which were used for training, validation, and testing the machine learning (ML) models. Data used in training the ML models include the well logs of density, porosity, electrical resistivity, gamma radiation, and acoustic wave velocity measurements. Combinations of two or three out of these five well logs were found to reliably predict the gas hydrate saturation with accuracy varying between 80 and 90% when compared to the gas hydrate saturations derived from Nuclear Magnetic Resonance (NMR)-based technique. The ML models trained on data from three ANS wells achieved high fidelity predictions of gas hydrate saturation at the Mallik site. The results obtained in this study indicate that ML models trained on data from one geological basin can successfully predict key reservoir parameters for permafrost-associated gas hydrate accumulations within another basin. A generalized approach for selecting a well log combination that can improve model accuracy is discussed. Overall, the study outcome supports earlier work demonstrating that ML models trained on non-NMR well logs are a viable alternative to physics-driven methods for predicting gas hydrate saturations.

58 GEOSCIENCES↗
DOE OSTI2024

Near-global summer circulation response to the spring surface temperature anomaly in Tibetan Plateau –– the GEWEX/LS4P first phase experiment

Subseasonal to seasonal (S2S) prediction of droughts and floods is one of the major challenges of weather and climate prediction. Recent studies suggest that the springtime land surface temperature/subsurface temperature (LST/SUBT) over the Tibetan Plateau (TP) can be a new source of S2S predictability. The project “Impact of Initialized Land Surface Temperature and Snowpack on Subseasonal to Seasonal Prediction (LS4P)” was initiated to study the impact of springtime LST/SUBT anomalies over high mountain areas on summertime precipitation predictions. The present work explores the simulated global scale response of the atmospheric circulation to the springtime TP land surface cooling by 16 current state-of-the-art Earth System Models (ESMs) participating in the LS4P Phase I (LS4P-I) experiment. The LS4P-I results show, for the first time, that springtime TP surface anomalies can modulate a persistent quasi-barotropic Tibetan Plateau-Rocky Mountain Circumglobal (TRC) wave train from the TP via the northeast Asia and Bering Strait to the western part of the North America, along with the springtime westerly jet from TP across the whole North Pacific basin. The TRC wave train modulated by the TP thermal anomaly play a critical role on the early summer surface air temperature and precipitation anomalies in the regions along the wave train, especially over the northwest North America and the southern Great Plains. The participant models that fail in capturing the TRC wave train greatly under-predict climate anomalies in reference to observations and the successful models. These results suggest that the TP LST/SUBT anomaly via the TRC wave train is the first order source of the S2S variability in the regions mentioned. Furthermore, the TP surface temperature anomaly can influence the Southern Hemispheric circulation by generating cross-equator wave trains. However, the simulated propagation pathways from the TP into the Southern Hemisphere show large inter-model differences. More dynamical understanding of the TRC wave train as well as its cross-equator propagation into the Southern Hemisphere will be explored in the newly launched LS4P phase II experiment.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Distributed Training and Optimization of Neural Networks

Deep learning models are yielding increasingly better performances thanks to multiple factors. To be successful, model may have large number of parameters or complex architectures and be trained on large dataset. This leads to large requirements on computing resource and turn around time, even more so when hyperparameter optimization is done (e.g. search over model architectures). While this is a challenge that goes beyond particle physics, we review the various ways to do the necessary computations in parallel, and put it in the context of high-energy physics.

Vlimant, Jean-Roch↗
DOE OSTI2024

Neural networks for estimation of divertor conditions in DIII-D using C III imaging

Deep learning approaches have been applied to images of C III emission in the lower divertor of DIII-D to develop models for estimating the level of detachment and magnetic configuration (X-point location and strike point radial location). The poloidal distance from the target to the C III emission front is used to represent the level of detachment. The models perform well on a test dataset not used in training, achieving $F_1$ scores as high as 0.99 for detachment state classification and root mean squared error (RMSE) as low as 2cm for front location regression. Predictions for shots with intermittent reattachment are studied, with class activation mapping used to aid in interpretation of the model predictions. Based on the success of these models, a third model was trained to predict the X-point location and strike point radial position from C III images. Though the dataset covers only a small range of possible magnetic configurations, the model shows promising results, achieving RMSE around 1cm for the test data.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Accelerating high-strain continuum-scale brittle fracture simulations with machine learning

Failure in brittle materials under dynamic loading conditions is a result of the propagation and coalescence of microcracks. Simulating this discrete crack evolution at the continuum level is computationally expensive or, in some cases, intractable, resulting in the need to make broad assumptions or neglect key physics. In this work, we have developed an approach using machine learning that overcomes the current inability to represent meso-scale physics at the macro-scale. Our approach leverages damage and stress data from a computationally expensive high-fidelity model that explicitly resolves microcrack behavior to build an inexpensive machine learning emulator. Once trained, the machine learning emulator is used to predict the evolution of crack length statistics, which then informs a continuum-scale constitutive model. This results in a significant speed-up of the workflow by four orders of magnitude. Both the machine learning emulator and the continuum-scale model are validated against the high-fidelity model and experimental data, respectively, showing excellent agreement. There are two key findings. The first is that we can reduce the dimensionality of the problem, establishing that the machine learning emulator only needs the length of the longest crack and one of the maximum stress components to capture the necessary physics. Another compelling finding is that the emulator can be trained in one experimental setting and transferred successfully to predict behavior in a different setting.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Bayesian sparse learning with preconditioned stochastic gradient MCMC and its applications

Deep neural networks have been successfully employed in an extensive variety of research areas, including solving partial differential equations. Despite its significant success, there are some challenges in effectively training DNN, such as avoiding overfitting in over-parameterized DNNs and accelerating the optimization in DNNs with pathological curvature. Here, we propose a Bayesian type sparse deep learning algorithm. The algorithm utilizes a set of spike-and-slab priors for the parameters in the deep neural network. The hierarchical Bayesian mixture will be trained using an adaptive empirical method. That is, one will alternatively sample from the posterior using preconditioned stochastic gradient Langevin Dynamics (PSGLD), and optimize the latent variables via stochastic approximation. The sparsity of the network is achieved while optimizing the hyperparameters with adaptive searching and penalizing. A popular SG-MCMC approach is Stochastic gradient Langevin dynamics (SGLD). However, considering the complex geometry in the model parameter space in nonconvex learning, updating parameters using a universal step size in each component as in SGLD may cause slow mixing. To address this issue, we apply a computationally manageable preconditioner in the updating rule, which provides a step-size parameter to adapt to local geometric properties. Moreover, by smoothly optimizing the hyperparameter in the preconditioning matrix, our proposed algorithm ensures a decreasing bias, which is introduced by ignoring the correction term in the preconditioned SGLD. According to the existing theoretical framework, we show that the proposed algorithm can asymptotically converge to the correct distribution with a controllable bias under mild conditions. Numerical tests are performed on both synthetic regression problems and learning solutions of elliptic PDE, which demonstrate the accuracy and efficiency of the present work.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

All-optical phase conjugation using diffractive wavefront processing

Abstract Optical phase conjugation (OPC) is a nonlinear technique used for counteracting wavefront distortions, with applications ranging from imaging to beam focusing. Here, we present a diffractive wavefront processor to approximate all-optical phase conjugation. Leveraging deep learning, a set of diffractive layers was optimized to all-optically process an arbitrary phase-aberrated input field, producing an output field with a phase distribution that is the conjugate of the input wave. We experimentally validated this wavefront processor by 3D-fabricating diffractive layers and performing OPC on phase distortions never seen during training. Employing terahertz radiation, our diffractive processor successfully performed OPC through a shallow volume that axially spans tens of wavelengths. We also created a diffractive phase-conjugate mirror by combining deep learning-optimized diffractive layers with a standard mirror. Given its compact, passive and multi-wavelength nature, this diffractive wavefront processor can be used for various applications, e.g., turbidity suppression and aberration correction across different spectral bands.

42 ENGINEERING↗
DOE OSTI2024

Optimizing Distributed Training on Frontier for Large Language Models

Large language models (LLMs) have demonstrated remarkable success as foundational models, benefiting various downstream applications through fine-tuning. Loss scaling studies have demonstrated the superior performance of larger LLMs compared to their smaller counterparts. Nevertheless, training LLMs with billions of parameters poses significant challenges and requires considerable computational resources. For example, training a one trillion parameter GPT-style model on 20 trillion tokens requires a staggering 120 million exaflops. This research explores efficient distributed training strategies to extract this computation from Frontier, the world's first exascale supercomputer. We enable and investigate various model and data parallel training techniques, such as tensor parallelism, pipeline parallelism, and sharded data parallelism, to facilitate training a trillion-parameter model on Frontier. We empirically assess these techniques and their associated parameters to determine their impact on memory footprint, communication latency, and GPU's computational efficiency. We analyze the complex interplay among these techniques and find a strategy to combine them to achieve high throughput through hyperparameter tuning. We have identified efficient strategies for training large LLMs of varying sizes through empirical analysis and hyperparameter tuning. For 22 Billion, 175 Billion, and 1 Trillion parameters, we achieved GPU throughputs of 38.38%, 36.14%, and 31.96%, respectively. For the training of the 175 Billion parameter model and the 1 Trillion parameter model, we achieved 100% weak scaling efficiency on 1024 and 3072 Mi250X GPUs, respectively. We also achieved strong scaling efficiencies of 89% and 87% for these two models. We trained these models only tens of iterations instead of training till completion.

Yin, Junqi↗
DOE OSTI2024

Predicting Mechanical Properties from Microstructure Images in Fiber-Reinforced Polymers Using Convolutional Neural Networks

Evaluating the mechanical response of fiber-reinforced composites can be extremely time-consuming and expensive. Machine learning (ML) techniques offer a means for faster predictions via models trained on existing input–output pairs and have exhibited success in composite research. This paper explores a fully convolutional neural network modified from StressNet, which was originally used for linear elastic materials, and extended here for a non-linear finite element (FE) simulation to predict the stress field in 2D slices of segmented tomography images of a fiber-reinforced polymer specimen. The network was trained and evaluated on data generated from the FE simulations of the exact microstructure. The testing results show that the trained network accurately captures the characteristics of the stress distribution, especially on fibers, solely from the segmented microstructure images. The trained model can make predictions within seconds in a single forward pass on an ordinary laptop, given the input microstructure, compared to 92.5 h to run the full FE simulation on a high-performance computing cluster. These results show promise in using ML techniques to conduct fast structural analysis for fiber-reinforced composites and suggest a corollary that the trained model can be used to identify the location of potential damage sites in fiber-reinforced polymers.

Sun, Yixuan (ORCID:0000000311093380)↗
DOE OSTI2022

Creating ground truth for nanocrystal morphology: a fully automated pipeline for unbiased transmission electron microscopy analysis

Control over colloidal nanocrystal morphology (size, size distribution, and shape) is important for tailoring the functionality of individual nanocrystals and their ensemble behavior. Despite this, traditional methods to quantify nanocrystal morphology are laborious. New developments in automated morphology classification will accelerate these analyses but the assessment of machine learning models is limited by human accuracy for ground truth, causing even unsupervised machine learning models to have inherent bias. Herein, we introduce synthetic image rendering to solve the ground truth problem of nanocrystal morphology classification. By simulating 2D images of nanocrystal shapes via a function of high-dimensional parameter space, we trained a convolutional neural network to link unique morphologies to their simulated parameters, defining nanocrystal morphology quantitatively rather than qualitatively. An automated pipeline then processes, quantitatively defines, and classifies nanocrystal morphology from experimental transmission electron microscopy (TEM) images. Using improved computer vision techniques, 42,650 nanocrystals were identified, assessed, and labeled with quantitative parameters, offering a 600-fold improvement in efficiency over best-practice manual measurements. Further, a classification algorithm was trained with a prediction accuracy of 99.5%, which can successfully analyze a range of concave, convex, and irregular nanocrystal shapes. The resulting pipeline was applied to differentiating two syntheses of nominally cuboidal CsPbBr 3 nanocrystals and uniquely classifying binary nickel sulfide nanocrystal phase based on morphology. This pipeline provides a simple, efficient, and unbiased method to quantify nanocrystal morphology and represents a practical route to construct large datasets with an absolute ground truth for training unbiased morphology-based machine learning algorithms.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2023

Machine Learning Enabled Sensor Fusion for In-Situ Defect Detection in Laser Powder Bed Fusion

Laser Powder Bed Fusion (L-PBF) Additive Manufacturing (AM) is among the metal 3D printing technologies most broadly adopted by the manufacturing industry. The current industry qualification paradigm for critical-application L-PBF parts relies heavily on expensive non-destructive inspection techniques such as X-Ray Computed Tomography (XCT), which significantly limits the use-cases of L-PBF. In situ monitoring of the process promises a less expensive alternative to ex situ testing, but existing sensor technologies and data analysis techniques struggle to detect sub-surface flaws (e.g., porosity and cracking) on production-scale L-PBF printers. RTX Technologies Research Center (RTRC) has licensed ORNL’s Peregrine software package – a printer- and camera-agnostic data analytics tool designed specifically for detecting process anomalies using in situ data collected during powder bed printing. The goal of this project was to feed temporally rich, multi-modal sensor data, including visible light, integrated near infrared (NIR), and spatially mapped co-axial melt pool thermal emission data into Peregrine to enable detection of subsurface flaws. XCT data was used as ground truth training data to allow Peregrine’s deep learning algorithms to recognize anomalies in these complex data streams in both test artifacts and industrially relevant geometries. Completion of this program has seen the successful implementation of multi-modal, multi-layer sensor data footprints for training of machine learning models in Peregrine. Flaws detected in XCT data have been successfully detected directly from this in situ data footprint, and initial analyses of the in situ probability-of-detection has been conducted, showing performance levels commensurate with traditional non-destructive evaluation (NDE) methods. The in situ monitoring methodology was then applied to an industrially relevant component that was using post-build NDE, highlighting the utility of the proposed method for hard-to-inspect AM components. As a direct result of this program, two journal manuscripts [1], [2] have been published in Additive Manufacturing, with additional manuscripts planned following program completion.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Bridging the Gap Between LLMs and LNS with Dynamic Data Format and Architecture Codesign

Deep neural networks (DNNs) have achieved tremendous success in the past few years. However, their training and inference demand exceptional computational and memory resources. Quantization has been shown as an effective approach to mitigate the cost, with the mainstream data types reduced from FP32 to FP16/BF16 and recently FP8 in the latest NVIDIA H100 GPUs. With increasingly aggressive quantization, however, the conventional floating-point formats suffer from limited precision in representing numbers around zero. Recently, NVIDIA demonstrated the potential of using a Logarithmic Number System (LNS) for the next generation of tensor cores. While LNS mitigates the hurdles in representing small numbers, in this work we observed a mismatch between LNS and the emerging Large Language Models (LLM), where LLM exhibits significant outliers when directly adopting the LNS format. In this paper, we present a data-format/architecture codesign to bright this gap. On the format side, we propose a dynamic LNS format to flexibly represent outliers at a higher precision, by exploiting asymmetry in the LNS representation and identifying outliers through a per-vector basis. On the architecture side, for demonstration, we realize the dynamic LNS format in a systolic array, which can handle the irregularity of the outliers at runtime. We implement our approach on an Alveo U280 FPGA as a prototype. Experimental results show that our design can effectively handle the outliers and resolve the mismatch between LNS and LLM, contributing to an accuracy improvement of 15.4% and 16% over the floating-point and the original LNS baselines, using four state-of-the-art LLM models. Our observation and design lay a solid foundation for the large-scale adoption of the LNS format in the next-generation deep learning hardware.

Haghi, Pouya↗
DOE OSTI2023

Streaming Data from Experimental Facilities to Supercomputers for Real-Time Data Processing

In this paper we demonstrate direct data streaming from instruments and detectors at a large-scale experimental facility to a supercomputer for real-time data processing and feedback. Streaming data to supercomputers introduces the potential for novel scientific applications and workflow models, including the ability to provide real-time feedback from very large datasets during an experiment and the integration of real-time ML training and inference at scale. We discuss a successful demonstration for real-time processing of data from the Advanced Photon Source (APS) on the Polaris supercomputer using an EPICS-based streaming framework. We describe the capabilities of the streaming framework itself, and outline the architecture that allows us to process experimentally derived data on a supercomputer without file-based data transfers. We present throughput measurements that are indicative of system performance capable of sustaining the expected data production rates of the facility, as well as discuss some outstanding challenges and our future directions.

real-time processing↗
DOE OSTI2020

SEDONA: SpEctroscopic Detection of Nerve Agents

Los Alamos National Laboratory developed the only portable screener to accurately detect the chemical nerve agents, providing results in seconds. Current airport detection system cannot scan for the threat of organophosphorus nerve agents and insecticides. Small quantities of these toxic chemicals can very quickly cause widespread exposure and severe symptoms. SEDONA screens through an unopened bottle using the principles of nuclear magnetic resonance spectroscopy. The system could be deployed and operated with minimal training. SEDONA dramatically reduces the likelihood of a successful nerve agent attack at airports, government buildings, embassies, sporting events, concerts, and political rallies.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

Building Monitoring and Energy Efficiency Training for Small and Mid-Sized Commercial Buildings in Non-Urbanized Areas of Alaska

The goal of this project was to create an effective training that increases understanding, competency and capacity in monitoring critical building data. Mentorship, verification and follow-up ensured participants now have necessary skills to install and maintain building monitoring systems. This training sought to empower building managers and maintenance staff to properly and efficiently operate their buildings for energy savings and increased lifespan of equipment. Through the project, participants’ knowledge of the building was increased through hands-on training using a familiar building environment and eliminating common barriers to effective training through peer-to peer delivery. In successful cases the host organization will save money and resources due to more efficient operation of building equipment already in place.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

An Extension Industrial Assessment Center in Eastern Great Lakes Region

The extension IAC at CWRU is funded for 3 years to conduct 10 assessments of small and medium-sized manufacturing facilities each year in Eastern Great Lakes region. Total 30 assessments was proposed in the SOPO but due to the Covid-19 pandemic, the IAC program was suspended for 1.5 years. A total of 20 assessments was contracted to the extension IAC at CWRU which have been successfully accomplished. In the proposal, we proposed to train 18 students in the extension IAC program, and actually we selected and trained 19 students even though the IAC project was suspended during half of its project period.

42 ENGINEERING↗
DOE OSTI2021

Estimating Watershed Subsurface Permeability From Stream Discharge Data Using Deep Neural Networks

Subsurface permeability is a key parameter in watershed models that controls the contribution from the subsurface flow to stream flows. Since the permeability is difficult and expensive to measure directly at the spatial extent and resolution required by fully distributed watershed models, estimation through inverse modeling has had a long history in subsurface hydrology. The wide availability of stream surface flow data, compared to groundwater monitoring data, provides a new data source to infer soil and geologic properties using integrated surface and subsurface hydrologic models. As most of the existing methods have shown difficulty in dealing with highly nonlinear inverse problems, we explore the use of deep neural networks for inversion owing to their successes in mapping complex, highly nonlinear relationships. We train various deep neural network (DNN) models with different architectures to predict subsurface permeability from stream discharge hydrograph at the watershed outlet. The training data are obtained from ensemble simulations of hydrographs corresponding to an permeability ensemble using a fully-distributed, integrated surface-subsurface hydrologic model. The trained model is then applied to estimate the permeability of the real watershed using its observed hydrograph at the outlet. Our study demonstrates that the permeabilities of the soil and geologic facies that make significant contributions to the outlet discharge can be more accurately estimated from the discharge data. Their estimations are also more robust with observation errors. Compared to the traditional ensemble smoother method, DNNs show stronger performance in capturing the nonlinear relationship between permeability and stream hydrograph to accurately estimate permeability. Our study sheds new light on the value of the emerging deep learning methods in assisting integrated watershed modeling by improving parameter estimation, which will eventually reduce the uncertainty in predictive watershed models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

An out-of-distribution-aware autoencoder model for reduced chemical kinetics

While detailed chemical kinetic models have been successful in representing rates of chemical reactions in continuum scale computational fluid dynamics (CFD) simulations, applying the models in simulations for engineering device conditions is computationally prohibitive. To reduce the cost, data-driven methods, e.g., autoencoders, have been used to construct reduced chemical kinetic models for CFD simulations. Despite their success, data-driven methods rely heavily on training data sets and can be unreliable when used in out-of-distribution (OOD) regions (i.e., when extrapolating outside of the training set). In this paper, we present an enhanced autoencoder model for combustion chemical kinetics with uncertainty quantification to enable the detection of model usage in OOD regions, and thereby creating an OOD-aware autoencoder model that contributes to more robust CFD simulations of reacting flows.Here, we first demonstrate the effectiveness of the method in OOD detection in two well-known datasets, MNIST and Fashion-MNIST, in comparison with the deep ensemble method, and then present the OOD-aware autoencoder for reduced chemistry model in syngas combustion.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗