Search NASA⌕ Search

SEARCH · Search NASA

Results for “network inference algorithms”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Data Summarization and Inference at Scale

This is the final report for the DOE ASCR grant SC-0022260, Data Summarization and Inference at Scale, PI: Alex Pothen, Purdue University. The goal of the project was to solve data-intensive and compute-intensive problems in the physical sciences, engineering, information science, data science, etc. by designing and implementing new algorithms that could work with a subset of the data. The four subgoals were: (a) The solution of problems where the data is too large to be stored in the memory of a computer. In this streaming model of computation, the data arrives as a stream of elements to the computer, each element is processed as it arrives, and a decision is made to discard the data or to store it; only a small subset of the data proportional to the size of the output solution is stored, and when all the data has been streamed, a solution to the problem is computed from the stored subset. (b) The use of machine learning methods to compute solutions to data-intensive problems. The use of GPUs is critical to obtain high performance on machine learning tasks, but their memory sizes are smaller relative to that of CPUs. For large-scale problems, the data is sampled many times, and small samples are used with repetition, for robustness, to compute solutions to inference tasks. This sampling reduces the memory required to solve the problem, but attention is needed to avoid slow convergence to the solutions, and reduced accuracy of inference. We propose submodular optimization, Large Language Models, and physics-informed neural networks to enable GPU computations here. (c) Modeling and visualization of high-dimensional data using interpretable features. Clinical proteomic data sets from immunology for the detection of cancer and other diseases are temporal and high-dimensional, and algorithms for visualizing these data sets using clinically interpretable features are lacking. We propose methods that compute distances based on the optimal transportation problem and graph edit distances to address this problem. We also propose the use of optimal transport-based distances, spatial statistics, and network structure to classify image data sets, We apply these algorithms to electron micrographs of the peripheral nervous system in the digestive tract. (d) The design of data-intensive algorithms on emerging architectures, specifically, noisy, intermediate-scale quantum (NISQ) devices. Quantum computers offer the possibility of exploring large solution spaces due to the principle of superposition, but current quantum computers are limited by few qubits, short coherence times due to noise, poor interconections among the qubits, etc. We propose the use of the divide and conquer paradigm to solve large-scale problems, wherein collections of small subproblems are solved on the quantum devices, and the solutions to the subproblems are integrated into a solution for the original problem on a classical computer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Inference of response functions with the help of machine-learning algorithms

Response functions are a key quantity to describe the near-equilibrium dynamics of strongly interacting many-body systems. Recent techniques that attempt to overcome the challenges of calculating these ab initio have employed expansions in terms of orthogonal polynomials. We employ a neural network prediction algorithm to reconstruct a response function 𝑆⁡(𝜔) defined over a range in frequencies 𝜔. Here, we represent the calculated response function as a truncated Chebyshev series whose coefficients can be optimized to reduce the representation error. We compare the quality of response functions obtained using coefficients calculated using a neural network (NN) algorithm with those computed using the Gaussian integral transform (GIT) method. In the regime where only a small number of terms in the Chebyshev series are retained, we find that the NN scheme outperforms the GIT method.

Kurkcuoglu, Doga Murat [Fermi National Accelerator↗
DOE OSTI2022

Inferring safety critical events from vehicle kinematics in naturalistic driving environment: Application of deep learning Algorithms

Advances in sensing technology has enabled the collection of countless terabytes of second-by-second kinematics data. Such data provides opportunities for real-time monitoring of driving behavior and identification of safety critical events (SCEs) including crashes and near crashes. The concept of volatility is relevant in this context, which identifies instability and erratic variations in driving behavior prior to involvement in SCEs. This study utilized vehicle kinematics from a large-scale naturalistic driving data to develop a deep learning approach based on 1D convolutional neural networks (CNN) for inferring SCEs. The data are unique in the sense that such accurate pre-crash data at high fidelity are not available in traditional crash repositories. This study contributes to the literature by providing a first attempt at predicting responses to SCEs by developing deep learning-based CNN architectures using novel driving volatility based kinematic thresholds for a sample of 9553 events. The key contribution lies in developing a volatility-based CNN input layout that is acceptable to CNN schemes and represents the motion kinematics such as speed, acceleration and volatility measures. Several 1D-CNN architectures were developed using layers, numbers of convolutions, layer patterns, and kernels. Shallow and deep architectures were tested, revealing higher accuracy of shallow architectures in detecting SCEs. The optimal number of epochs were identified using an early stopping method while the CNN performance was improved by increasing the number of epochs. The ensemble CNN had the highest predictive accuracy of 95.6% for detection of crashes and near crashes, which was 2.5% higher than the optimal CNN using 20% hold out test data. The ensemble CNN also outperformed classical machine learning models and model performance reported in past studies on detection of SCEs. Finally, these results have implications for identification of safety hotspots and providing real-time alerts and warnings in connected and highly automated vehicle environment including society of automotive engineers levels 3–5.

42 ENGINEERING↗
DOE OSTI2023

The influence of physical and algorithmic factors on simulated far-field waveforms and source–time functions of underground explosions using unsupervised machine learning

SUMMARY Characterizing explosion sources and differentiating between earthquake and underground explosions using distributed seismic networks becomes non-trivial when explosions are detonated in cavities or heterogeneous ground material. Moreover, there is little understanding of how changes in subsurface physical properties affect the far-field waveforms we record and use to infer information about the source. Simulations of underground explosions and the resultant ground motions can be a powerful tool to systematically explore how different subsurface properties affect far-field waveform features, but there are added variables that arise from how we choose to model the explosions that can confound interpretation. To assess how both subsurface properties and algorithmic choices affect the seismic wavefield and the estimated source functions, we ran a series of 2-D axisymmetric non-linear numerical explosion experiments and wave propagation simulations that explore a wide array of parameters. We then inverted the synthetic far-field waveform data using a linear inversion scheme to estimate source–time functions (STFs) for each simulation case. We applied principal component analysis (PCA), an unsupervised machine learning method, to both the far-field waveforms and STFs to identify the most important factors that control variance in the waveform data and differences between cases. For the far-field waveforms, the largest variance occurs in the shallower radial receiver channels in the 0–50 Hz frequency band. For the STFs, both peak amplitude and rise times across different frequencies contribute to the variance. We find that the ground equation of state (i.e. lithology and rheology) and the explosion emplacement conditions (i.e. tamped versus cavity) have the greatest effect on the variance of the far-field waveforms and STFs, with the ground yield strength and fracture pressure being secondary factors. Differences in the PCA results between the far-field waveforms and STFs could possibly be due to near-field non-linearities of the source that are not accounted for in the estimation of STFs and could be associated with yield strength, fracture pressure, cavity radius and cavity shape parameters. Other algorithmic parameters are found to be less important and cause less variance in both the far-field waveforms and STFs, meaning algorithmic choices in how we model explosions are less important, which is encouraging for the further use of explosion simulations to study how physical Earth properties affect seismic waveform features and estimated STFs.

58 GEOSCIENCES↗
DOE OSTI2023

Semi-supervised graph neural networks for pileup noise removal

Abstract The high instantaneous luminosity of the CERN Large Hadron Collider leads to multiple proton–proton interactions in the same or nearby bunch crossings (pileup). Advanced pileup mitigation algorithms are designed to remove this noise from pileup particles and improve the performance of crucial physics observables. This study implements a semi-supervised graph neural network for particle-level pileup noise removal, by identifying individual particles produced from pileup. The graph neural network is firstly trained on charged particles with known labels, which can be obtained from detector measurements on data or simulation, and then inferred on neutral particles for which such labels are missing. This semi-supervised approach does not depend on the neutral particle pileup label information from simulation, and thus allows us to perform training directly on experimental data. The performance of this approach is found to be consistently better than widely-used domain algorithms and comparable to the fully-supervised training using simulation truth information. The study serves as the first attempt at applying semi-supervised learning techniques to pileup mitigation, and opens up a new direction of fully data-driven machine learning pileup mitigation studies.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Inferring fracture dilation and shear slip from surface deformation utilising trained surrogate models

An important task in energy and CO 2 storage (sequestration) in the subsurface is to verify that the surrounding fractures and faults are not activated, acting as leakage pathways. This is achievable through effective and efficient Measurement, Monitoring and Verification (MMV) plans. In this work, two surrogate models are trained to captures dilation (opening) and shear deformation of fractures, and the associated surface deformation. The trained surrogate model, based on conditional Generative-Adversarial Networks (cGAN) receives fracture apertures from dilational fractures together with fracture slips from shear fractures and predicts the combined surface deformation. An inversion algorithm based on Bayesian framework is proposed to identify the geometry of both types of fractures, as well as volume of dilational fractures and deformation moment induced by shear fractures, all from the measured surface deformation data. The inversion algorithm utilises the Differential Evolution (DE) optimisation technique that has the superior performance in finding the global minimum of cost function. The proposed surrogate-assisted inversion successfully inferred the unknown dip, dip direction and the volume of the dilational fractures as well as the induced deformation moment in shear fractures. The model was further tested for the inversion of a field hydraulic fracturing tilt dataset applying different scenarios with varying unknowns to show the model's performance, as well as incorporating shear deformation for better match with the observed data.

Dilation and shear↗
DOE OSTI2021

Integration of a DER Management System in Riverside. Final report

The tasks in this project covered various aspects, including algorithm development, algorithm integration into a commercial Active Network Management (ANM) platform, hardware-in-the-loop (HIL) testing in an industry-standard testing platform, pilot demonstration in Riverside, California, and also cost and benefit analysis. The DERMS platform in this project can host different algorithms developed on different platforms (e.g., MATLAB and Python) and it can interact with different hardware devices (e.g., different PV inverters, battery inverters, and different sensors). The DER control solution are based on an advanced model-free, layered, and clustered DER control paradigm. At the core of the DER control algorithms was the concept of Extremum Seeking (ES), which is a model-free probing-based control technique. The ES-based control algorithms were tested on major real-world inverters; both individually and in a cluster. It was shown that even legacy equipment (or when paired with a few additional advanced equipment) can support such advanced control. The monitoring algorithms utilize a heterogeneous set of legacy and advanced sensor measurements, such as behind-the-meter DER sensors, distribution-level Phase Measurement Units, distribution-substation Supervisory Control and Data Acquisition (SCADA), and line current sensors, with their limited availability; in order to infer practical network conditions. Sensor data are utilized to achieve resource forecasting, phase identification, and distribution system state estimation. The technology that was developed and demonstrated in this project could be transformational to utilities, including the smaller municipal utilities such as in Riverside, which may not have the resources to deploy advanced distribution system and DERMS solutions in order to support high penetration of solar power integration. This project created a real-world prototype to provide utilities with an assessment of smart grid monitoring and control technologies.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS2022

SatNet: A Benchmark for Satellite Scheduling Optimization

Satellites provide essential services such as networking and weather tracking, and the number of near-earth and deep space satellites are expected to grow rapidly in the coming years. Communications with terrestrial ground stations is one of the critical functionalities of any space mission. Satellite scheduling is a problem that has been scientifically investigated since the 1970s. A central aspect of this problem is the need to consider resource contention and satellite visibility constraints as they require line of sight. Due to the combinatorial nature of the problem, prior solutions such as linear programs and evolutionary algorithms require extensive compute capabilities to output a feasible schedule for each scenario. Machine learning based scheduling can provide an alternative solution by training a model with historical data and generating a schedule quickly with model inference. We present SatNet, a benchmark for satellite scheduling optimization based on historical data from the NASA Deep Space Network. We propose formulation of the satellite scheduling problem as a Markov Decision Process and use reinforcement learning (RL) policies to generate schedules. The nature of constraints imposed by SatNet differ from other combinatorial optimization problems such as vehicle routing studied in prior literature. Our initial results indicate that RL is an alternative optimization approach that can generate candidate solutions of comparable quality to existing state-of-the-practice results. However, we also find that RL policies overfit to the training dataset and do not generalize well to new data, thereby necessitating continued research on reusable and generalizable agents.

Wilson, Brian↗
DOE OSTI2022

Probabilistic Nanomagnetic Memories for Uncertain and Robust Machine Learning

This project evaluated the use of emerging spintronic memory devices for robust and efficient variational inference schemes. Variational inference (VI) schemes, which constrain the distribution for each weight to be a Gaussian distribution with a mean and standard deviation, are a tractable method for calculating posterior distributions of weights in a Bayesian neural network such that this neural network can also be trained using the powerful backpropagation algorithm. Our project focuses on domain-wall magnetic tunnel junctions (DW-MTJs), a powerful multi-functional spintronic synapse design that can achieve low power switching while also opening the pathway towards repeatable, analog operation using fabricated notches. Our initial efforts to employ DW-MTJs as an all-in-one stochastic synapse with both a mean and standard deviation didn’t end up meeting the quality metrics for hardware-friendly VI. In the future, new device stacks and methods for expressive anisotropy modification may make this idea still possible. However, as a fall back that immediately satisfies our requirements, we invented and detailed how the combination of a DW-MTJ synapse encoding the mean and a probabilistic Bayes-MTJ device, programmed via a ferroelectric or ionically modifiable layer, can robustly and expressively implement VI. This design includes a physics-informed small circuit model, that was scaled up to perform and demonstrate rigorous uncertainty quantification applications, up to and including small convolutional networks on a grayscale image classification task, and larger (Residual) networks implementing multi-channel image classification. Lastly, as these results and ideas all depend upon the idea of an inference application where weights (spintronic memory states) remain non-volatile, the retention of these synapses for the notched case was further interrogated. These investigations revealed and emphasized the importance of both notch geometry and anisotropy modification in order to further enhance the endurance of written spintronic states. In the near future, these results will be mapped to effective predictions for room temperature and elevated operation DW-MTJ memory retention, and experimentally verified when devices become available.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2020

Radiation Source Localization Using Surrogate Models Constructed from 3-D Monte Carlo Transport Physics Simulations

Recent research has focused on the development of surrogate models for radiation source localization in a simulated urban domain. We employ the Monte Carlo N-Particle (MCNP) code to provide high- delity simulations of radiation transport within an urban domain. The model is constructed to employ a source location (x, y, z) as input and return the estimated count rate for a set of speci ed detector locations. Because MCNP simulations are computationally expensive, we develop e cient and accurate surrogate models of the detector responses. We construct surrogate models using Gaussian processes (GP) and neural networks (NN) that we train and verify using the MCNP simulations. The trained surrogate models provide an e cient framework for Bayesian inference and experimental design. We employ Delayed Rejection Adaptive Metropolis (DRAM), a Markov Chain Monte Carlo (MCMC) algorithm, to infer the location and intensity of an unknown source. The DRAM results yield a posterior probability distribution for the source's location conditioned on the observed detector count rates. The posterior distribution exhibits regions of high and low probability within the simulated environment identifying potential source locations. In this manner, we can quantify the source location to within at least one of these regions of high probability in the considered cases. Employing these methods, we are able to reduce the space of potential source locations by at least 60%.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2026

Wilson loops with neural networks

Wilson loops are essential objects in QCD and have been pivotal in scale setting and demonstrating confinement. Various generalizations are crucial for computations needed in effective field theories. In lattice gauge theory, Wilson loop calculations face challenges, including excited-state contamination at short times and the signal-to-noise ratio issue at longer times. To address these problems, we develop a new method by using neural networks to parametrize interpolators for the static quark-antiquark pair. We construct gauge-equivariant layers for the network and train it to find the ground state of the system. The trained network itself is then treated as our new observable for the inference. Our results demonstrate a significant improvement in the signal compared to traditional Wilson loops, performing as well as Coulomb-gauge Wilson-line correlators while maintaining gauge invariance. Additionally, we present an example where the optimized ground state is used to measure the static force directly, as well as another example combining this method with the multilevel algorithm. Finally, we extend the formalism to find excited-state interpolators for static quark-antiquark systems. To our knowledge, this work is the first study of neural networks with a physically motivated loss function for Wilson loops.

Bellscheidt, Verena [Massachusetts Inst. of Techno↗
DOE OSTI2023

Accelerating Machine Learning Inference with GPUs in ProtoDUNE Data Processing

Abstract We study the performance of a cloud-based GPU-accelerated inference server to speed up event reconstruction in neutrino data batch jobs. Using detector data from the ProtoDUNE experiment and employing the standard DUNE grid job submission tools, we attempt to reprocess the data by running several thousand concurrent grid jobs, a rate we expect to be typical of current and future neutrino physics experiments. We process most of the dataset with the GPU version of our processing algorithm and the remainder with the CPU version for timing comparisons. We find that a 100-GPU cloud-based server is able to easily meet the processing demand, and that using the GPU version of the event processing algorithm is two times faster than processing these data with the CPU version when comparing to the newest CPUs in our sample. The amount of data transferred to the inference server during the GPU runs can overwhelm even the highest-bandwidth network switches, however, unless care is taken to observe network facility limits or otherwise distribute the jobs to multiple sites. We discuss the lessons learned from this processing campaign and several avenues for future improvements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

BitGNN: Unlocking the Performance Potential of Binary Graph Neural Networks on GPUs

Graph Neural Networks (GNNs) have shown compelling results in many graph-based learning tasks. They are, however, time-consuming. Recent work has shown a promising direction in improving GNN speed and shrinking the size — network binarization, which binarizes network values and operations. Prior work, however, mainly focused on algorithm designs, leaving it open on how to fully materialize the performance potential. This work fills the gap by proposing techniques to best map binary GNNs and their computations to fit the nature of bit manipulations, optimizations and algorithms to maximize BSpMM kernel efficiency, and solutions to other factors influencing the end-to-end time on GPUs. Results on real-world graphs show that the proposed techniques outperform state of-the-art binary GNN implementations by 21-67× with little accuracy loss.

Chen, Jou-An↗
DOE OSTI2025

Analog Systems for Edge Optimization

Over the past decade, analog computing has the subject of substantial research interest providing a path toward improved computational efficiency in the post-Dennard era. Analog matrix vector multiplication (MVM) accelerators provide a popular approach given the ubiquity of MVM operations in numerous applications. However, historically analog computing systems can struggle with applications requiring high precision due to the inherent susceptibility of these systems to analog non-idealities. Therefore, prior work on analog systems has focused either on applications known to be tolerant of limited precision (e.g., neural network inference), or using expensive techniques to emulate high-precision using many analog MVM operations. In this work, we propose an alternative approach. Motivated by recent advances in inexact nonlinear solvers and optimizers, we explore the potential of co-designing optimization algorithms which can take full advantage of the fundamentally inexact analog MVM operations. To enable these co-designed algorithms we also develop a general mathematical theory of the precision and energy efficiency of analog operations, and a new system architecture for tightly-coupled analog and digital computation. Finally, we examine the applicability of analog computing to a wider class of symmetric positive definite systems and find potential in using analog operations as a sparse approximate inverse preconditioner. With these core innovations, this project provides a path toward effectively implementing optimization algorithms on power-constrained autonomous and semi-autonomous systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

WHISPER: Wireless Home Identification and Sensing Platform for Energy Reduction

Many regions of the world benefit from heating, ventilating, and air-conditioning (HVAC) systems to provide productive, comfortable, and healthy indoor environments, which are enabled by automatic building controls. Due to climate change, population growth, and industrialization, HVAC use is globally on the rise. Unfortunately, these systems often operate in a continuous fashion without regard to actual human presence, leading to unnecessary energy consumption. As a result, the heating, ventilation, and cooling of unoccupied building spaces makes a substantial contribution to the harmful environmental impacts associated with carbon-based electric power generation, which is important to remedy. For our modern electric power system, transitioning to low-carbon renewable energy is facilitated by integration with distributed energy resources. Automatic engagement between the grid and consumers will be necessary to enable a clean yet stable electric grid, when integrating these variable and uncertain renewable energy sources. We present the WHISPER (Wireless Home Identification and Sensing Platform for Energy Reduction) system to address the energy and power demand triggered by human presence in homes. The presented system includes a maintenance-free and privacy-preserving human occupancy detection system wherein a local wireless network of battery-free environmental, acoustic energy, and image sensors are deployed to monitor homes, record empirical data for a range of monitored modalities, and transmit it to a base station. Several machine learning algorithms are implemented at the base station to infer human presence based on the received data, harnessing a hierarchical sensor fusion algorithm. Results from the prototype system demonstrate an accuracy in human presence detection in excess of 95%; ongoing commercialization efforts suggest approximately 99% accuracy. Using machine learning, WHISPER enables various applications based on its binary occupancy prediction, allowing situation-specific controls targeted at both personalized smart home and electric grid modernization opportunities.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS2021

Extensive Characterization of Aerosol Optical Properties and Chemical Component Concentrations: Application of the GRASP/Component Approach to Long-term AERONET Measurements

A recently developed GRASP/Component approach (GRASP: Generalized Retrieval of Atmosphere and Surface Properties) was applied to AERONET (Aeronet Robotic Network) sun photometer measurements in this study. Unlike traditional aerosol component retrieval, this approach allows the inference of some information about aerosol composition directly from measured radiance, rather than indirectly through the inversion of optical parameters, and has been integrated into the GRASP algorithm. The newly developed GRASP/Component approach was applied to 13 AERONET sites for different aerosol types under the assumption of aerosol internal mixing rules to analyze the characteristics of aerosol components and their distribution patterns. The results indicate that the retrievals can characterize well the spatial and temporal variability of the component concentration for different aerosol types. A reasonable agreement between GRASP BC retrievals and MERRA-2 BC products is found for all different aerosol types. In addition, the relationships between aerosol component content and aerosol optical parameters such as aerosol optical depth (AOD), fine-mode fraction (FMF), absorption Ångström exponent (AAE), scattering Ångström exponent (SAE), and single scattering albedo (SSA) are also analyzed for indirect verifying the reliability of the component retrieval. It was demonstrated the GRASP/Component optical retrievals are in good agreement with AERONET standard products [e.g., correlation coefficient (R) of 0.93–1.0 for AOD, fine-mode AOD (AODF), coarse-mode AOD (AODC) and Ångström exponent (AE); R = ~ 0.8 for absorption AOD (AAOD) and SSA; RMSE (root mean square error) < 0.03 for AOD, AODF, AODC, AAOD and SSA]. Thus, it is demonstrated the GRASP/Component approach can provide aerosol optical products with comparable accuracy as the AERONET standard products from the ground-based sun photometer measurements as well as some additional important inside on aerosol composition.

Xindan Zhang↗
DOE OSTI2022

Data reduction through optimized scalar quantization for more compact neural networks

Raw data generation for several existing and planned large physics experiments now exceeds TB/s rates, generating untenable data sets in very little time. Those data often demonstrate high dimensionality while containing limited information. Meanwhile, Machine Learning algorithms are now becoming an essential part of data processing and data analysis. Those algorithms can be used offline for post processing and post data analysis, or they can be used online for real time processing providing ultra low latency experiment monitoring. Both use cases would benefit from data throughput reduction while preserving relevant information: one by reducing the offline storage requirements by several orders of magnitude and the other by allowing ultra fast online inferencing with low complexity Machine Learning models. Moreover, reducing the data source throughput also reduces material cost, power and data management requirements. In this work we demonstrate optimized nonuniform scalar quantization for data source reduction. This data reduction allows lower dimensional representations while preserving the relevant information of the data, thus enabling high accuracy Tiny Machine Learning classifier models for online fast inferences. We demonstrate this approach with an initial proof of concept targeting the CookieBox, an array of electron spectrometers used for angular streaking, that was developed for LCLS-II as an online beam diagnostic tool. We used the Lloyd-Max algorithm with the CookieBox dataset to design an optimized nonuniform scalar quantizer. Optimized quantization lets us reduce input data volume by 69% with no significant impact on inference accuracy. When we tolerate a 2% loss on inference accuracy, we achieved 81% of input data reduction. Finally, the change from a 7-bit to a 3-bit input data quantization reduces our neural network size by 38%.

97 MATHEMATICS AND COMPUTING↗