Search NASA⌕ Search

SEARCH · Search NASA

Results for “sparse learning”

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 523 records · Page 29

Scaled Vecchia Approximation for Fast Computer-Model Emulation

Many scientific phenomena are studied using computer experiments consisting of multiple runs of a computer model while varying the input settings. Gaussian processes (GPs) are a popular tool for the analysis of computer experiments, enabling interpolation between input settings, but direct GP inference is computationally infeasible for large datasets. We adapt and extend a powerful class of GP methods from spatial statistics to enable the scalable analysis and emulation of large computer experiments. Specifically, we apply Vecchia’s ordered conditional approximation in a transformed input space, with each input scaled according to how strongly it relates to the computer-model response. The scaling is learned from the data by estimating parameters in the GP covariance function using Fisher scoring. Our methods are highly scalable, enabling estimation, joint prediction, and simulation in near-linear time in the number of model runs. In several numerical examples, our approach substantially outperformed existing methods.

97 MATHEMATICS AND COMPUTING↗

How Low Can You Go? Using Synthetic 3D Imagery to Drastically Reduce Real-World Training Data for Object Detection

Deep convolutional neural networks (DCNNs) currently provide state-of-the-art performance on image classification and object detection tasks, and there are many global security mission areas where such models could be extremely useful. Crucially, the success of these models is driven in large part by the widespread availability of high-quality open source data sets such as Image Net, Common Objects in Context (COCO), and KITTI, which contain millions of images with thousands of unique labels. However, global security relevant objects-of-interest can be difficult to obtain: relevant events are low frequency and high consequence; the content of relevant images is sensitive; and adversaries and proliferators seek to obscure their activities. For these cases where exemplar data is hard to come-by, even fine-tuning an existing model with available data can be effectively impossible. Recent work demonstrated that models can be trained using a combination of real-world and synthetic images generated from 3D representations; that such models can exceed the performance of models trained using real-world data alone; and that the generated images need not be perfectly realistic (Tremblay, et al., 2018). However, this approach still required hundreds to thousands of real-world images for training and fine tuning, which for sparse, global security-relevant datasets can be an unrealistic hurdle. In this research, we validate the performance and behavior of DCNN models as we drive the number of real-world images used for training object detection tasks down to a minimal set. We perform multiple experiments to identify the best approach to train DCNNs from an extremely small set of real-world images. In doing so, we: Develop state-of-the-art, parameterized 3D models based on real-world images and sample from their parameters to increase the variance in synthetic image training data; Use machine learning explainability techniques to highlight and correct through targeted training the biases that result from training using completely synthetic images; and Validate our results by comparing the performance of the models trained on synthetic data to one another, and to a control model created by fine-tuning an existing ImageNet-trained model with a limited number (hundreds) of real-world images.

97 MATHEMATICS AND COMPUTING↗

Radiochemistry and the Study of Fission

These are slides from a lecture given at UC Berkeley. Radiochemistry has been used to study fission since its discovery. Radiochemical methods are used to determine cumulative mass yields. These measurements have led to the two-mode fission hypothesis to model the neutron energy dependence of fission product yields. Fission product yields can be used for the nuclear forensics of nuclear explosions. The mass yield curve depends on both the fuel and the neutron spectrum of a device. Recent studies have shown that the nuclear structure of the compound nucleus can affect the mass yield distribution. The following topics are covered: In the beginning: the discovery of fission; forensics using fission products: what can be learned from fission products, definitions of R-values and Q-values, fission bases, K-factors and fission chambers, limitations; the neutron energy dependence of the mass yield distribution (the two mode fission hypothesis); the influence of nuclear structure on the mass yield distribution. In summary: Radiochemistry has been used to study fission since its discovery. Radiochemical measurement of fission product yields have provided the highest precision data for developing fission models and for nuclear forensics. The two-mode fission hypothesis provides a description of the neutron energy dependence of the mass yield curve. However, data is still rather sparse and more work is needed near second and third chance fission. Radiochemical measurements have provided evidence for the importance of nuclear states in the compound nucleus in predicting the mass yield curve in the resonance region.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗

Binning sequences using very sparse labels within a metagenome

Background: In metagenomic studies, a process called binning is necessary to assign contigs that belong to multiple species to their respective phylogenetic groups. Most of the current methods of binning, such as BLAST, k-mer and PhyloPythia, involve assigning sequence fragments by comparing sequence similarity or sequence composition with already-sequenced genomes that are still far from comprehensive. We propose a semi-supervised seeding method for binning that does not depend on knowledge of completed genomes. Instead, it extracts the flanking sequences of highly conserved 16S rRNA from the metagenome and uses them as seeds (labels) to assign other reads based on their compositional similarity. Results: The proposed seeding method is implemented on an unsupervised Growing Self-Organising Map (GSOM), and called Seeded GSOM (S-GSOM). We compared it with four well-known semi-supervised learning methods in a preliminary test, separating random-length prokaryotic sequence fragments sampled from the NCBI genome database. We identified the flanking sequences of the highly conserved 16S rRNA as suitable seeds that could be used to group the sequence fragments according to their species. S-GSOM showed superior performance compared to the semi-supervised methods tested. Additionally, S-GSOM may also be used to visually identify some species that do not have seeds. The proposed method was then applied to simulated metagenomic datasets using two different confidence threshold settings and compared with PhyloPythia, k-mer and BLAST. At the reference taxonomic level Order, S-GSOM outperformed all k-mer and BLAST results and showed comparable results with PhyloPythia for each of the corresponding confidence settings, where S-GSOM performed better than PhyloPythia in the ≥ 10 reads datasets and comparable in the ≥ 8 kb benchmark tests. Conclusion: In the task of binning using semi-supervised learning methods, results indicate S-GSOM to be the best of the methods tested. Most importantly, the proposed method does not require knowledge from known genomes and uses only very few labels (one per species is sufficient in most cases), which are extracted from the metagenome itself. These advantages make it a very attractive binning method. SGSOM outperformed the binning methods that depend on already-sequenced genomes, and compares well to the current most advanced binning method, PhyloPythia.

59 BASIC BIOLOGICAL SCIENCES↗

Multi-static Serial LiDAR for Surveillance and Identification of Marine Life at MHK Installations

Final Report for project DE-EE0006787: Multi-static Serial LiDAR for Surveillance and Identification of Marine Life at MHK Installations. This project developed and tested an optical monitoring system prototype that will be suitable for marine and hydrokinetic (MHK) full project lifecycle observation (baseline, commissioning, and decommissioning), with automated real-time classification of marine animals. This system can be deployed to collect pre-installation baseline species observations at a proposed deployment site with minimal post-processing overhead. To satisfy deployed MHK project species of concern (e.g. Endangered Species Act-listed) monitoring requirements, the system provides automated tracking and notification of the presence of managed animals within established perimeters of MHK equipment and provides high resolution imagery of their behavior through a wide range of conditions. During a project’s decommissioning stage, the system can remain installed to provide resource managers with post-installation data. Our technology, known as an Unobtrusive Multi-static Serial LiDAR Imager (UMSLI), is a technology transfer of underwater distributed LiDAR imaging technology that preserves the advantages of traditional optical and acoustic solutions while overcoming associated disadvantages for MHK environmental monitoring applications. This new approach is a purposefully-designed, reconfigurable adaptation of an existing technology that can be easily mounted on or around different classes of MHK equipment. The system uses low average power red (638nm) laser illumination to be invisible and eye-safe to marine animals and is compact and cost effective. The equipment is designed for long term, maintenance-free operations, to inherently generate a sparse primary dataset that only includes detected anomalies (animal presence information), and to allow robust real-time automated animal classification/identification with a low data bandwidth requirement. Advantages of the technology over others currently being used or being considered for MHK monitoring include: Unlike a conventional camera, the depth of field is near-infinite and limited by attenuation (approximately 5-8 m) rather than focal properties of a lens; Operation in an adaptive mode which can project a sparse grid of pulses with higher peak power for longer range detection (>10 meters) and track animals within a zone of interest with high resolution imagery for identification of marine life at closer range (<5m); System detection limit and Signal-to-Noise-Ratio is superior to a camera, due to rejection of both backscattering component and ambient solar background; Multiple wide-angle pulsed laser illuminators and bucket detectors can be flexibly configured to cover a 4pi steradian (i.e. omnidirectional) scene volume, while also retrieving 3D features of animal targets from timing information; Process and classification framework centered around a novel active learning and incremental classification classifier that enables accurate identification of a variety of marine animals automatically; A two-tiered monitoring architecture and invisible watermarking-based data archiving and retrieving approach ensures significant data reduction while preserving high fidelity monitoring. A methodology to train and optimize the classifier for target species of concern to optimize site monitoring effectiveness. This technological innovation addresses a high priority regulatory requirement to observe marine life interaction near MHK projects. Our solution improves resource manager confidence that any interactions between marine animals and equipment are observed in a cost-effective and automated manner. Without EERE funding, this novel application of multi-static LiDAR would not have been available to the MHK community for environmental monitoring.

16 TIDAL AND WAVE POWER↗

Stochastic projective splitting

Here, we present a new, stochastic variant of the projective splitting (PS) family of algorithms for inclusion problems involving the sum of any finite number of maximal monotone operators. This new variant uses a stochastic oracle to evaluate one of the operators, which is assumed to be Lipschitz continuous, and (deterministic) resolvents to process the remaining operators. Our proposal is the first version of PS with such stochastic capabilities. We envision the primary application being machine learning (ML) problems, with the method’s stochastic features facilitating “mini-batch” sampling of datasets. Since it uses a monotone operator formulation, the method can handle not only Lipschitz-smooth loss minimization, but also min–max and noncooperative game formulations, with better convergence properties than the gradient descent-ascent methods commonly applied in such settings. The proposed method can handle any number of constraints and nonsmooth regularizers via projection and proximal operators. We prove almost-sure convergence of the iterates to a solution and a convergence rate result for the expected residual, and close with numerical experiments on a distributionally robust sparse logistic regression problem.

97 MATHEMATICS AND COMPUTING↗

A regularization approach to hydrofacies delineation

We consider an inverse problem of identifying complex internal structures of composite (geological) materials from sparse measurements of system parameters and system states. Two conceptual frameworks for identifying internal boundaries between constitutive materials in a composite are considered. A sequential approach relies on support vector machines, nearest neighbor classifiers, or geostatistics to reconstruct boundaries from measurements of system parameters and then uses system states data to refine the reconstruction. A joint approach inverts the two data sets simultaneously by employing a regularization approach.

58 GEOSCIENCES↗

Validation and Sensitivity Analysis of a 1-D Lake Model Across Global Lakes

Lakes have important influence on weather and climate from local to global scales. However, their prediction using numerical models is notoriously difficult because lakes are highly heterogeneous across the globe, but observations are sparse. In this study, we assessed the performance of a 1-D lake model in simulating the thermal structures of 58 lakes with diverse morphometric and geographic characteristics by following the phase 2a local lake protocol of the Inter-sectoral Impact Model Intercomparison Project (ISIMIP2a). After calibration, the root-mean-square errors (RMSE) were below 2 °C for 70% and 75% of the lakes for epilimnion and full-profile temperature simulations, with an average of 1.71 °C and 1.43 °C, respectively. The model performance mainly depended on lake shape rather than location, supporting the possibility of grouping model parameters by lake shape for global applications. Furthermore, through machine-learning based parameter sensitivity tests, we identified turbulent heat fluxes, wind-driven mixing and water transparency as the major processes controlling lake thermal and mixing regimes. Snow density was also important for modeling the ice phenology of high-latitude lakes. The relative influence of the key processes and the corresponding parameters mainly depended on lake latitude and depth. Turbulent heat fluxes showed a decreasing importance in affecting epilimnion temperature with increasing latitude. Wind-driven mixing was less influential to lake stratification for deeper lakes while the impact of light extinction, on the contrary, showed a positive correlation with depth. Our findings may guide improvements in 1-D lake model parameterizations to achieve higher fidelity in simulating global lake thermal dynamics.

54 ENVIRONMENTAL SCIENCES↗

Variance Preserving Spectral Subsampling

Generating statistically faithful short-duration gamma-ray spectra from a single long measurement is essential in nuclear safeguards, supporting tasks such as algorithm development and machine-learning applications, especially when list-mode data are unavailable. Existing subsampling methods often distort the statistical characteristics of genuine short-duration measurements, leading to biased or unreliable analytical outcomes and thereby undermining downstream tasks. In this work, we compare five subsampling approaches using a benchmark set of 156 genuine replicate spectra collected with a high-purity germanium detector. We evaluate each method with respect to run-to-run variance, channel-to-channel variance, and preservation of total counts (losslessness). Across a wide range of subsampling ratios, only binomial subsampling without replacement consistently reproduces the statistical properties of genuine short-duration spectra, maintaining proper dispersion even in sparse spectral regions and perfectly preserving total counts. These results provide a mathematically principled and practically validated framework for generating synthetically shortened spectra when true short-duration measurements are unavailable.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

Using Multiple Geophysical Methods to Refine a Stratigraphic Conceptual Site Model at a Nuclear Waste Site

At contaminated sites, there is a critical reliance on conceptual site models of which a key component is the geologic framework model (GFM), which is used to build predictive fate and transport modeling in support of remediation goals. Here, we detail an ongoing study at a nuclear waste site (Hanford Site, Washington, USA) where surface geophysical methods were used to complement existing borehole information and to help site new boreholes to update and refine the GFM. Ground-based electrical resistivity tomography (ERT), time-domain electromagnetics (TEM), and seismic data was acquired on the Hanford Site’s Central Plateau, where the presence of a hydraulically transmissive subsurface feature, or paleochannel, was suspected but uncertain based on contaminant concentrations at sparsely located boreholes. To assess and compare ERT, TEM, and seismic geophysical datasets, co-located data was acquired in an area where the existing GFM was more certain. Surrounding the suspected paleochannel, ERT and seismic refraction tomography showed similar subsurface structures consistent with a channelized feature. While TEM interpretation was limited in the suspected paleochannel area due to a thick resistive top layer and high electromagnetic noise, TEM showed greater utility in a different area where these factors were less prevalent. Based on the results of quasi-3D ERT inversions, we propose updates to the GFM, including identifying a paleochannel. Further, we evaluate the utility of geophysical methods and examine lessons learned that will be used as a basis for continued characterization efforts to support site management decisions and implementation efforts.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Coupled machine learning–ecosystem ensemble models substantially improve predictions of nitrous oxide (N 2 O) fluxes from US croplands

Nitrous oxide (N 2 O) is a potent and persistent greenhouse gas, with rising atmospheric concentrations driven in part by inefficient use of synthetic nitrogen (N) fertilizers in agriculture. Predicting soil N 2 O emissions is challenging due to high spatial and temporal variability arising from complex soil biogeochemical processes. Process-based ecosystem models and standalone machine learning (ML) approaches without extensive site-specific calibration often miss high-emission episodes. Here, we show how an Ensemble Modeling System (EMS) based on outputs from an ensemble of ecosystem models coupled to an ensemble of ML models can improve predictions and understanding of N 2 O fluxes from US cropland. Trained and validated on ~12,000 N 2 O chamber measurements at 17 US Midwest sites (six crops, 35 management practices), the EMS accurately predicted daily fluxes of N 2 O at both training (R 2 = 0.84, RMSE = 16.4 g N ha −1 d −1 ) and held-out testing sites (R 2 = 0.84, RMSE = 6.2 g N ha −1 d −1 ). Analyses identified six dominant N 2 O drivers: soil organic carbon (SOC), NH 4 + , NO 3 - , water-filled pore space, temperature, and aboveground biomass production. Wet, warm soils produced large N 2 O peaks only with sufficient SOC and mineral N; in low-SOC soils, fluxes remained low. Incorporating these drivers into process-based models might significantly improve their predictive capacity. The EMS demonstrates a strong potential to predict N 2 O fluxes at unseen sites, enabling more reliable regional inventories, improved gap-filling where measurements are sparse, and enhanced understanding of mechanisms to advance targeted mitigation strategies in food, feed, and bioenergy crops.

AI↗

Extreme-Scale Algorithms & Software Resilience (EASIR) Architecture-Aware Algorithms for Scalable Performance and Resilience on Heterogeneous Architectures

This project addresses both communication-avoiding algorithms, and reproducible floating-point computation. Communication, i.e. moving data, either between levels of memory or processors over a network, is much more expensive per operation than arithmetic (measured in time or energy), so we seek algorithms that greatly reduce communication. We developed many new algorithms for both dense and sparse, and both direct and iterative linear algebra, attaining new communication lower bounds, and getting large speedups in many cases. We also extended this work in several ways: (1) We minimize writes separately from reads, since writes may be much more expensive than reads on emerging memory technologies, like Flash, sometimes doing asymptotically fewer writes than reads. (2) We extend the lower bounds and optimal algorithms to arbitrary algorithms that may be expressed as perfectly nested loops accessing arrays, where the array subscripts may be arbitrary affine functions of the loop indices (eg A(i), B(i,j+k, k+3*m-7, …) etc.). (3) We extend our communication-avoiding approach to some machine learning algorithms, such as support vector machines. This work has won a number of awards. We also address reproducible floating-point computation. We define reproducibility to mean getting bitwise identical results from multiple runs of the same program, perhaps with different hardware resources or other changes that should ideally not change the answer. Many users depend on reproducibility for debugging or correctness. However, dynamic scheduling of parallel computing resources, combined with nonassociativity of floating point addition, makes attaining reproducibility a challenge even for simple operations like summing a vector of numbers, or more complicated operations like the Basic Linear Algebra Subprograms (BLAS). We describe an algorithm that computes a reproducible sum of floating point numbers, independent of the order of summation. The algorithm depends only on a subset of the IEEE Floating Point Standard 754-2008, uses just 6 words to represent a “reproducible accumulator,” and requires just one read-only pass over the data, or one reduction in parallel. New instructions based on this work are being considered for inclusion in the future IEEE 754-2018 floating-point standard, and new reproducible BLAS are being considered for the next version of the BLAS standard.

97 MATHEMATICS AND COMPUTING↗

Improving Reliability of Large Language Models for Nuclear Power Plant Diagnostics Technical Presentation

Large Language Models (LLMs) struggle out of the box when answering factually about detailed questions, especially in domains that are sparsely represented in their training data. This causes hallucinations and reduces reliability making it difficult for them to be used in practice. This work shows that using RAG techniques can improve factual accuracy and reliability, allowing for the application of LLMs in specialized areas, even when those areas that aren’t extensively covered in their initial training.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

AWB-GCN: A Graph Convolutional Network Accelerator with Runtime Workload Rebalancing

The recent development of deep learning has been mostly focusing on Euclidean data, such as images, videos, audios, etc. However, most real-world information and relation are often expressed as graphs. To efficiently learn from graph data, graph convolutional networks (GCNs) emerge as a promising approach, showing advantages in several practical applications such as social network analysis, knowledge discovery, 3D modeling, motion capturing, etc. Real-world graphs are usually extremely large and imbalanced, posting significant performance demand and design challenges on the hardware dedicated for GCN inference. In this paper, we propose an architecture design called UW-GCN to accelerate graph convolutional network inference. To tackle the major performance bottleneck from workload imbalance, we propose dynamic neighborhood stealing and remote chunk shuffling techniques, relying on hardware flexibility to achieve hardware auto-tuning under negligible area or delay overhead. Specifically, UW-GCN is able to smartly profile the sparse graph pattern while continuously adjusting the workload distribution via routing reconfiguration among parallel processing elements (PEs). The ideal configuration is then reused in the remaining iterations. To the best of our knowledge, this is the first accelerator design particularly for GCN and the first work relying on hardware auto-tuning, which is normally based on software, to achieve near-optimal workload balance in processing sparse structures.

Geng, Tong↗

The Kokkos OpenMPTarget Backend: Implementation and Lessons Learned

As the supercomputing landscape diversifies, solutions such as Kokkos to write vendor agnostic applications and libraries have risen in popularity. Kokkos provides a programming model designed for performance portability, which allows developers to write a single source implementation that can run efficiently on various architectures. At its heart, Kokkos maps parallel algorithms to architecture and vendor specific backends written in lower level programming models such as CUDA and HIP. Another approach to writing vendor agnostic parallel code is using OpenMP’s directives based approach, which lets developers annotate code to express parallelism. It is implemented at the compiler level and is supported by all major high performance computing vendors, as well as the primary Open Source toolchains GNU and LLVM. Since its inception, Kokkos has used OpenMP to parallelize on CPU architectures. In this paper, we explore leveraging OpenMP for a GPU backend and discuss the challenges we encountered when mapping the Kokkos APIs and semantics to OpenMP target constructs. As an exemplar workload we chose a simple conjugate gradient solver for sparse matrices. We find that performance on NVIDIA and AMD GPUs varies widely based on details of the implementation strategy and the chosen compiler. Furthermore, the performance of the OpenMP implementations decreases with increasing complexity of the investigated algorithms.

Gayatri, Rahulkumar↗

Improving Reliability of Large Language Models for Nuclear Power Plant Diagnostics [Poster]

Large Language Models (LLMs) struggle out of the box when answering factually about detailed questions, especially in domains that are sparsely represented in their training data. This causes hallucinations and reduces reliability making it difficult for them to be used in practice. This work shows that using RAG techniques can improve factual accuracy and reliability, allowing for the application of LLMs in specialized areas, even when those areas that aren’t extensively covered in their initial training.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Improved Subseasonal Forecasting of Extreme Polar Vortices Using Machine Learning

Our research was focused on forecasting the position and shape of the winter stratospheric polar vortex at a subseasonal timescale of 15 days in advance. To achieve this, we employed both statistical and neural network machine learning techniques. The analysis was performed on 42 winter seasons of reanalysis data provided by NASA giving us a total of 6,342 days of data. The state of the polar vortex for determined by using geometric moments to calculate the centroid latitude and the aspect ratio of an ellipse fit onto the vortex. Timeseries for thirty additional precursors were calculated to help improve the predictive capabilities of the algorithm. Feature importance of these precursors was performed using random forest to measure the predictive importance and the ideal number of precursors. Then, using the precursors identified as important, various statistical methods were tested for predictive accuracy with random forest and nearest neighbor performing the best. An echo state network, a type of recurrent neural network that features sparsely connected hidden layer and a reduced number of trainable parameters that allows for rapid training and testing, was also implemented for the forecasting problem. Hyperparameter tuning was performed for each methods using a subset of the training data. The algorithms were trained and tuned on the first 41 years of data, then tested for accuracy on the final year. In general, the centroid latitude of the polar vortex proved easier to predict than the aspect ratio across all algorithms. Random forest outperformed other statistical forecasting algorithms overall but struggled to predict extreme values. Forecasting from echo state network suggested a strong predictive capability past 15 days, but further work is required to fully realize the potential of recurrent neural network approaches.

54 ENVIRONMENTAL SCIENCES↗

Regularized machine learning on molecular graph model explains systematic error in DFT enthalpies

Abstract A major goal of materials research is the discovery of novel and efficient heterogeneous catalysts for various chemical processes. In such studies, the candidate catalyst material is modeled using tens to thousands of chemical species and elementary reactions. Density Functional Theory (DFT) is widely used to calculate the thermochemistry of these species which might be surface species or gas-phase molecules. The use of an approximate exchange correlation functional in the DFT framework introduces an important source of error in such models. This is especially true in the calculation of gas phase molecules whose thermochemistry is calculated using the same planewave basis set as the rest of the surface mechanism. Unfortunately, the nature and magnitude of these errors is unknown for most practical molecules. Here, we investigate the error in the enthalpy of formation for 1676 gaseous species using two different DFT levels of theory and the ‘ground truth values’ obtained from the NIST database. We featurize molecules using graph theory. We use a regularized algorithm to discover a sparse model of the error and identify important molecular fragments that drive this error. The model is robust to rigorous statistical tests and is used to correct DFT thermochemistry, achieving more than an order of magnitude improvement.

36 MATERIALS SCIENCE↗