Search NASA⌕ Search

SEARCH · Search NASA

Results for “compressed 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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Elucidating size effects on the yield strength of single-crystal Cu via the Richtmyer–Meshkov instability

Capturing the dynamic response of a material under high strain-rate deformation often demands challenging and time consuming experimental effort. While shock hydrodynamic simulation methods can aid in this area, a priori characterizations of the material strength under shock loading and spall failure are needed in order to parameterize constitutive models needed for these computational tools. Moreover, parameterizations of strain-rate-dependent strength models are needed to capture the full suite of Richtmyer–Meshkov instability (RMI) behavior of shock compressed metals, creating an unrealistic demand for these training data solely on experiments. Herein, we sweep a large range of geometric, crystallographic, and shock conditions within molecular dynamics (MD) simulations and demonstrate the breadth of RMI in Cu that can be captured from the atomic scale. In this work, yield strength measurements from jetted and arrested material from a sinusoidal surface perturbation were quantified as Y RMI = 0.787 ± 0.374 GPa, higher than strain-rate-independent models used in experimentally matched hydrodynamic simulations. Defect-free, single-crystal Cu samples used in MD will overestimate Y RMI , but the drastic scale difference between experiment and MD is highlighted by high confidence neighborhood clustering predictions of RMI characterizations, yielding incorrect classifications.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

AutoAtlas: Neural Network for 3D Unsupervised Partitioning and Representation Learning

Here we present a novel neural network architecture called AutoAtlas for fully unsupervised partitioning and representation learning of 3D brain Magnetic Resonance Imaging (MRI) volumes. AutoAtlas consists of two neural network components: one neural network to perform multi-label partitioning based on local texture in the volume, and a second neural network to compress the information contained within each partition. We train both of these components simultaneously by optimizing a loss function that is designed to promote accurate reconstruction of each partition, while encouraging spatially smooth and contiguous partitioning, and discouraging relatively small partitions. We show that the partitions adapt to the subject specific structural variations of brain tissue while consistently appearing at similar spatial locations across subjects. AutoAtlas also produces very low dimensional features that represent local texture of each partition. We demonstrate prediction of metadata associated with each subject using the derived feature representations and compare the results to prediction using features derived from FreeSurfer anatomical parcellation. Since our features are intrinsically linked to distinct partitions, we can then map values of interest, such as partition-specific feature importance scores onto the brain for visualization.

42 ENGINEERING↗
DOE OSTI2021

Robustness of deep learning algorithms in astronomy -- galaxy morphology studies

Deep learning models are being increasingly adopted in wide array of scientific domains, especially to handle high-dimensionality and volume of the scientific data. However, these models tend to be brittle due to their complexity and overparametrization, especially to the inadvertent adversarial perturbations that can appear due to common image processing such as compression or blurring that are often seen with real scientific data. It is crucial to understand this brittleness and develop models robust to these adversarial perturbations. To this end, we study the effect of observational noise from the exposure time, as well as the worst case scenario of a one-pixel attack as a proxy for compression or telescope errors on performance of ResNet18 trained to distinguish between galaxies of different morphologies in LSST mock data. We also explore how domain adaptation techniques can help improve model robustness in case of this type of naturally occurring attacks and help scientists build more trustworthy and stable models.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Evaluation of Drilling Performance at The Geysers with Machine Learning Methods Using Geologic Data

A recent well, GDC-36, was drilled in The Geysers Geothermal Field served in a Department of Energy-industry to demonstrate improved drilling performance with polycrystalline diamond compact (PDC) bits. Both PDC and roller cone drill bits were used to drill this well. Key challenges encountered during drilling included lost circulation in the mud-drilled section, and bit damage interfacial severity in the deeper, air-drilled section. The objective of this study is to evaluate the drilling performance in relation to the local geological characteristics using machine learning methods. By applying K-clustering to the sonic log data, we were able to identify areas correlated with measured lost circulation. Also, the boundaries defined by clustering of the mineralogical and lithological data from the mud logs correlate well with interfacial severity during drilling. A random forest model was employed to build correlation between drilling data and rock strength. The confined compressive strength (CCS) of the rock in the training of the machine learning model was inferred from the dipole sonic log. The R-squared of the testing data is 0.78, and the RMSE (Root Mean Squared Error) is 0.06. The trained model was used to forecast rock strength for the section where sonic log data are not available. CCS could also be inferred from mud logs provided the relationship between mineralogy and rock strength is established through core testing data.

15 GEOTHERMAL ENERGY↗
DOE OSTI2022

Nonnegative canonical tensor decomposition with linear constraints: nnCANDELINC

Abstract There is an emerging interest for tensor factorization applications in big‐data analytics and machine learning. To speed up the factorization of extra‐large datasets, organized in multidimensional arrays (also known as tensors), easy to compute compression‐based tensor representations, such as, Tucker and tensor train formats, are used to approximate the initial large‐tensor. Further, tensor factorization is used to extract latent features that can facilitate discoveries of new mechanisms and signatures hidden in the data, where the explainability of the latent features is of principal importance. Nonnegative tensor factorization extracts latent features that are naturally sparse and parts of the data, which makes them easily interpretable. However, to take into account available domain knowledge and subject matter expertise, often additional constraints need to be imposed, which lead us to canonical decomposition with linear constraints (CANDELINC), a canonical polyadic decomposition with rank deficient factors. In CANDELINC, Tucker compression is used as a preprocessing step, which lead to a larger residual error but to more explainable latent features. Here, we propose a nonnegative CANDELINC (nnCANDELINC) accomplished via a specific nonnegative Tucker decomposition; we refer to as minimal or canonical nonnegative Tucker. We derive several results required to understand the specificity of nnCANDELINC, focusing on the difficulties of preserving the nonnegative rank of a tensor to its Tucker core and comparing the real valued to nonnegative case. Finally, we demonstrate nnCANDELINC performance on synthetic and real‐world examples.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Direct interpolative construction of the discrete Fourier transform as a matrix product operator

The quantum Fourier transform (QFT), which can be viewed as a reindexing of the discrete Fourier transform (DFT), has been shown to be compressible as a low-rank matrix product operator (MPO) or quantized tensor train (QTT) operator. However, the original proof of this fact does not furnish a construction of the MPO with a guaranteed error bound. Meanwhile, the existing practical construction of this MPO, based on the compression of a quantum circuit, is not as efficient as possible. We present a simple closed-form construction of the QFT MPO using the interpolative decomposition, with guaranteed near-optimal compression error for a given rank. This construction can speed up the application of the QFT and the DFT, respectively, in quantum circuit simulations and QTT applications. We also connect our interpolative construction to the approximate quantum Fourier transform (AQFT) by demonstrating that the AQFT can be viewed as an MPO constructed using a different interpolation scheme.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Utilizing computer vision and artificial intelligence algorithms to predict and design the mechanical compression response of direct ink write 3D printed foam replacement structures

Additive Manufacturing (AM) of porous polymeric materials, such as foams, recently became a topic of intensive research due their unique combination of low density, impressive mechanical properties, and stress dissipation capabilities. Conventional methods for fabricating foams rely on complex and stochastic processes, making it challenging to achieve precise architectural control of structured porosity. In contrast, AM provides access to a wide range of printable materials, where precise spatial control over structured porosity can be modulated during the fabrication process enabling the production of foam replacement structures (FRS). Current approaches for designing FRS are based on intuitive understanding of their properties or an extensive number of finite element method (FEM) simulations. These approaches, however, are computationally expensive and time consuming. As such, in this work, we present a novel methodology for determining the mechanical compression response of direct ink write (DIW) 3D printed FRS using a simple cross-sectional image. By obtaining measurement data for a relatively small number of samples, an artificial neural network (ANN) was trained, and a computer vision algorithm was used to make inferences about foam compression characteristics from a single cross-sectional image. Finally, a genetic algorithm (GA) was used to solve the inverse design problem, generating the AM printing parameters that an engineer should use to achieve a desired compression response from a DIW printed FRS. The methods developed herein present an avenue for entirely autonomous design and analysis of additively manufactured structures using artificial intelligence.

36 MATERIALS SCIENCE↗
DOE OSTI2025

A Variational Autoencoder Model Toward Molecular Structure Representation Learning of Fuels

Here, in this work, a Variational Autoencoder (VAE)-based data-driven modeling framework is developed with the overarching goal of enabling fuel design. The VAE model is trained on a large dataset with several chemical species to learn a compressed latent space molecular representation. Chemical structure in the form of Simplified Molecular Input Line Entry System (SMILES) string is fed as input, encoded into the VAE latent space, and decoded back to the SMILES string using Long Short-Term Memory (LSTM) networks. Complexities of the VAE training loss function are thoroughly examined by varying the weightage (beta (𝜷) parameter) of the latent space regularization term, thereby assessing the balance between reconstruction accuracy and validity, and focusing on both accurate molecular structure reconstruction and latent space consistency. Two different strategies for 𝜷 variation are evaluated: linear annealing and cyclic annealing. In addition, the impact of total correlation adjustment and hierarchical priors is also studied with regard to the balance between reconstruction fidelity and latent space regularization, and potential issues such as posterior collapse, over-regularization, and poor disentanglement of latent variables. Overall, the best performance of the model is achieved with hierarchical priors and incrementally increasing 𝜷 from 0 to a threshold value of 0.25 over 75 epochs. The generative VAE model can be readily coupled with Quantitative Structure–Property Relationship (QSPR) analysis to develop an integrated end-to-end framework for fuel-property prediction and molecular design of novel promising fuels.

fuel design↗
DOE OSTI2021

Fast convolutional neural networks on FPGAs with hls4ml

We introduce an automated tool for deploying ultra low-latency, low-power deep neural networks with convolutional layers on field-programmable gate arrays (FPGAs). By extending the hls4ml library, we demonstrate an inference latency of 5 µs using convolutional architectures, targeting microsecond latency applications like those at the CERN Large Hadron Collider. Considering benchmark models trained on the Street View House Numbers Dataset, we demonstrate various methods for model compression in order to fit the computational constraints of a typical FPGA device used in trigger and data acquisition systems of particle detectors. In particular, we discuss pruning and quantization-aware training, and demonstrate how resource utilization can be significantly reduced with little to no loss in model accuracy. We show that the FPGA critical resource consumption can be reduced by 97% with zero loss in model accuracy, and by 99% when tolerating a 6% accuracy degradation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2026

Surrogate Neural Architecture Codesign Package (SNAC-Pack)

Neural architecture search (NAS) is a powerful approach for automating model design, but existing methods often optimize for accuracy alone or rely on proxy metrics such as bit operations (BOPs) that correlate poorly with hardware cost. This gap is particularly large for FPGA deployment, where cost is dominated by a multi-dimensional budget of lookup tables, DSPs, flip-flops, BRAM, and latency. We present the Surrogate Neural Architecture Codesign Package (SNAC-Pack), an open-source AutoML framework for hardware-aware neural architecture codesign and end-to-end FPGA deployment. SNAC-Pack runs a multi-objective global search with Optuna and NSGA-II, loading trials to a shared SQLite store that enables parallel workers across compute nodes. A hardware surrogate model outputs per-trial resource and latency estimates, avoiding the synthesis cost that would otherwise dominate the search loop. A local search stage then applies quantization-aware training (QAT) together with iterative magnitude pruning in a combined compression loop, after which the final model is synthesized to FPGA firmware via the hls4ml Python library. A YAML configuration and an optional agentic frontend let users run the pipeline on new datasets without modifying the framework. We demonstrate SNAC-Pack on jet classification at the Large Hadron Collider and superconducting qubit readout, discovering compact architectures that match or exceed strong baselines on the task metric while reducing FPGA resource utilization and, in the qubit readout case, reducing the design space exploration process from months of manual fine-tuning to hours of automated search.

Weitz, Jason [UC, San Diego]↗
DOE OSTI2025

Dynamical Low-Rank Compression of Neural Networks with Robustness under Adversarial Attacks

Deployment of neural networks on resource-constrained devices demands models that are both compact and robust to adversarial inputs. However, compression and adversarial robustness often conflict. In this work, we introduce a dynamical low-rank training scheme enhanced with a novel spectral regularizer that controls the condition number of the low-rank core in each layer. This approach mitigates the sensitivity of compressed models to adversarial perturbations without sacrificing clean accuracy. The method is model- and data-agnostic, computationally efficient, and supports rank adaptivity to automatically compress the network at hand. Extensive experiments across standard architectures, datasets, and adversarial attacks show the regularized networks can achieve over 94 compression while recovering or improving adversarial accuracy relative to uncompressed baselines.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗
DOE OSTI2021

Visualization of shocked material instabilities using a fast-framing camera and XFEL four-pulse train

Many questions regarding dynamic materials could be answered by using time-resolved ultra-fast imaging techniques to characterize the physical and chemical behavior of materials in extreme conditions and their evolution on the nanosecond scale. In this work, we perform multi-frame phase-contrast imaging (PCI) of micro-voids in low density polymers under laser-driven shock compression. At the Matter in Extreme Conditions (MEC) Instrument at the Linac Coherent Light Source (LCLS), we used a train of four x-ray free electron laser (XFEL) pulses to probe the evolution of the samples. To visualize the void and shock wave interaction, here, we deployed the Icarus V2 detector to record up to four XFEL pulses, separated by 1-3 nanoseconds. In this work, we image elastic waves interacting with the micro-voids at a pressure of several GPa. Monitoring how the material’s heterogeneities, like micro-voids, dictate its response to a compressive wave is important for benchmarking the performances of inertial confinement fusion energy materials. For the first time in a single sample, we have combined an ultrafast x-ray framing camera and four XFEL pulse train to create an ultrafast movie of micro-void evolution under laser-driven shock compression. Eventually, we hope this technique will resolve the material density as it evolves dynamically under laser shock compression.

fusion↗
DOE OSTI2025

FedEFsz: Fair Cross-Silo Federated Learning System With Error-Bounded Lossy Compression

Cross-Silo federated learning systems have been identified as an efficient approach to scaling DNN training across geographically-distributed data silos to preserve the privacy of the training data. Communication efficiency and fairness are two major issues that need to be both satisfied when federated learning systems are deployed in practice. Simultaneously guaranteeing both of them, however, is exceptionally difficult because simply combining communication reduction and fairness optimization approaches often causes non-converged training or drastic accuracy degradation. Here, to bridge this gap, we propose FedEFsz. On the one hand, it integrates the state-of-the-art error-bounded lossy compressor SZ3 into cross-silo federated learning systems to significantly reduce communication traffic during the training. On the other hand, it achieves a high fairness (i.e., rather consistent model accuracy and performance across different clients) through a carefully designed heuristic algorithm that can tune the error-bound of SZ3 for different clients during the training. Extensive experimental results based on a GPU cluster with 65 GPU cards show that FedEFsz improves the fairness across different benchmarks by up to 60.88% and meanwhile reduces the communication traffic by up to 315×.

Cross-Silo Federated Learning Systems↗
DOE OSTI2023

A robust estimator of mutual information for deep learning interpretability

Abstract We develop the use of mutual information (MI), a well-established metric in information theory, to interpret the inner workings of deep learning (DL) models. To accurately estimate MI from a finite number of samples, we present GMM-MI (pronounced ‘Jimmie’), an algorithm based on Gaussian mixture models that can be applied to both discrete and continuous settings. GMM-MI is computationally efficient, robust to the choice of hyperparameters and provides the uncertainty on the MI estimate due to the finite sample size. We extensively validate GMM-MI on toy data for which the ground truth MI is known, comparing its performance against established MI estimators. We then demonstrate the use of our MI estimator in the context of representation learning, working with synthetic data and physical datasets describing highly non-linear processes. We train DL models to encode high-dimensional data within a meaningful compressed (latent) representation, and use GMM-MI to quantify both the level of disentanglement between the latent variables, and their association with relevant physical quantities, thus unlocking the interpretability of the latent representation. We make GMM-MI publicly available in this GitHub repository.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2020

Hands-on Introduction to Shock Physics (2020 Report, RR2020-IPD-1)

Los Alamos National Laboratory (LANL) researchers have played key roles in the development of high-pressure science and shock physics since the 1950s. Central to the Laboratory’s experimental capability are multiple impact systems and explosive loading capabilities that span impact velocities in excess of 8 km/s. More recently, our efforts have taken us to 3rd and 4th generation light sources where these loading platforms are coupled to X-ray diagnostics that are now providing new insights into matter-at-extremes. As the availability of these new diagnostics and loading techniques increases, there is an increased need within the high pressure community for researchers with the knowledge base and expertise to use them. Few institutions offer any level of theoretical or experimental coursework designed to train students for work in the field of shock physics or dynamic compression science, and those that do cannot meet the current demand of the national laboratories.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

DNTTD (Distributed Non-Negative Tensor Train Decomposition)

The era of exascale computing opens new venues for innovations and discoveries in many scientific, engineering, and commercial fields. However, with the exa flops also come the extra-large high-dimensional data generated by high performance computing. High-dimensional data is presented as multidimensional arrays, aka tensors. The presence of latent (not directly observable) structures in the tensor allows a unique representation and compression of the data by classical tensor factorization techniques. However, the classical tensor methods are not always stable or they can be exponential in their memory requirements, which makes them not suitable for high-dimensional tensors. Tensor train (TT) is a state-of-the-art tensor network introduced for factorization of high-dimensional tensors. TT transforms the initial high-dimensional tensor in a network of three-dimensional tensors that requires only a linear storage. Many real-world data, such as, density, temperature, population, probability, etc., are non-negative and for an easy interpretation, the algorithms preserving non-negativity are preferred. Here, we introduce a distributed non-negative tensor-train and demonstrate its scalability and the compression on synthetic and real world big datasets.

Bhattarai, Manish↗
DOE OSTI2024

Diamond precipitation dynamics from hydrocarbons at icy planet interior conditions

The pressure and temperature conditions at which precipitation of diamond occurs from hydrocarbon mixtures is important for modelling the interior dynamics of icy planets. However, there is substantial disagreement from laboratory experiments, with those using dynamic compression techniques finding much more extreme conditions are required than in static compression. Here we report the time-resolved observation of diamond formation from statically compressed polystyrene, (C 8 H 8 ) n , heated using the 4.5 MHz X-ray pulse trains at the European X-ray Free Electron Laser facility. Diamond formation is observed above 2,500 K from 19 GPa to 27 GPa, conditions representative of Uranus’s and Neptune’s shallow interiors, on 30 μs to 40 μs timescales. This is much slower than may be observed during the ~10 ns duration of typical dynamic compression experiments, revealing reaction kinetics to be the reason for the discrepancy. In conclusion, reduced pressure and temperature conditions for diamond formation has implications for icy planetary interiors, where diamond subduction leads to heating and could drive convection in the conductive ice layer that has a role in their magnetic fields.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Bridging material models across scales: An integrated approach to equation of state and molecular dynamics modeling of copper

New uncertainty-aware equation of state (EOS) and electrical conductivity (EC) models for copper have been developed. The multiphase EOS/EC models are fit to experimental solid/liquid EC isobar measurements as well as density-functional theory molecular dynamics (DFT-MD) EC calculations in both expanded and compressed regimes (0.1–16 g/ cm 3 ⁠). The liquid and solid EOS phases were fit to available experimental data along with additional DFT-MD data over the same range as the EC. Leveraging the DFT-MD data, a corresponding machine-learned interatomic potential (MLIAP) for copper was trained using genetic-algorithm optimization. The copper MLIAP was constrained by EOS shock points at high compressions. The final EOS bounded MLIAP proves to be stable over a large density range (approximately 0.1–20 g/ cm 3 ) with good agreement to an isothermal compression curve, shock Hugoniot, and liquid speed of sound measurements at high pressures (100s of GPa).

Acoustic measurements and instrumentation↗