Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “space networking”

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 1,315 records · Page 73
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Effects of Pore Connectivity on the Sorption of Fluids in Nanoporous Material: Ethane and CO2 Sorption in Silicalite

Adsorption of fluids in nanoporous materials is important for several applications including gas storage and catalysis. The pore network in natural, as well as engineered, materials can exhibit different degrees of connectivity between pores. While this might have important implications for the sorption of fluids, the effects of pore connectivity are seldom addressed in the studies of fluid sorption. We have carried out Monte Carlo simulations of the sorption of ethane and CO2 in silicalite, a nanoporous material characterized by sub-nanometer pores of different geometries (straight and zigzag channel like pores), with varied degrees of pore connectivity. The variation in pore connectivity is achieved by selectively blocking some pores by loading them with methane molecules that are treated as a part of the rigid nanoporous matrix in the simulations. Normalized to the pore space available for adsorption, the magnitude of sorption increases with a decrease in pore connectivity. The increased adsorption in the systems where pore connections are removed by blocking them is because of additional, albeit weaker, adsorption sites provided by the blocker molecules. By selectively blocking all straight or zigzag channels, we find differences in the absorption behavior of guest molecules in these channels.

Gautam, Siddharth (ORCID:0000000314435382)↗
DOE OSTI2021

Neutron Radiography and Computed Tomography of Biological Systems at the Oak Ridge National Laboratory's High Flux Isotope Reactor

Neutrons have historically been used for a broad range of biological applications employing techniques such as small-angle neutron scattering, neutron spin echo, diffraction, and inelastic scattering. Unlike neutron scattering techniques that obtain information in reciprocal space, attenuation-based neutron imaging measures a signal in real space that is resolved on the order of tens of micrometers. The principle of neutron imaging follows the Beer-Lambert law and is based on the measurement of the bulk neutron attenuation through a sample. Greater attenuation is exhibited by some light elements (most notably, hydrogen), which are major components of biological samples. Contrast agents such as deuterium, gadolinium, or lithium compounds can be used to enhance contrast in a similar fashion as it is done in medical imaging, including techniques such as optical imaging, magnetic resonance imaging, X-ray, and positron emission tomography. For biological systems, neutron radiography and computed tomography have increasingly been used to investigate the complexity of the underground plant root network, its interaction with soils, and the dynamics of water flux in situ. Moreover, efforts to understand contrast details in animal samples, such as soft tissues and bones, have been explored. This manuscript focuses on the advances in neutron bioimaging such as sample preparation, instrumentation, data acquisition strategy, and data analysis using the High Flux Isotope Reactor CG-1D neutron imaging beamline. The aforementioned capabilities will be illustrated using a selection of examples in plant physiology (herbaceous plant/root/soil system) and biomedical applications (rat femur and mouse lung).

Bilheux, Hassina↗
DOE OSTI2025

SchrödingerNet: A Universal Neural Network Solver for the Schrödinger Equation

Recent advances in machine learning have facilitated numerically accurate solution of the electronic Schrödinger equation (SE) by integrating various neural network (NN)-based wave function ansatzes with variational Monte Carlo methods. Nevertheless, such NN-based methods are all based on the Born–Oppenheimer approximation (BOA) and require computationally expensive training for each nuclear configuration. In this work, we propose a novel NN architecture, SchrödingerNet, to solve the full electronic-nuclear SE by defining a loss function designed to equalize local energies across the system. This approach is based on a translationally, rotationally and permutationally symmetry-adapted total wave function ansatz that includes both nuclear and electronic coordinates. Furthermore, this strategy not only allows for an efficient and accurate generation of a continuous potential energy surface at any geometry within the well-sampled nuclear configuration space, but also incorporates non-BOA corrections, through a single training process. Comparison with benchmarks of atomic and small molecular systems demonstrates its accuracy and efficiency.

Chemical calculations↗
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↗
DOE OSTI1991

Nuclear Safety [Vol. 32, No. 1, January-March 1991]

Nuclear Safety is a review journal that covers significant developments in the field of nuclear safety. Its scope includes the analysis and control of hazards associated with nuclear energy, operations involving fissionable materials, and the products of nuclear fission and their effects on the environment. Primary emphasis is on safety in reactor design, construction, and operation; however, the safety aspects of the entire fuel cycle, including fuel fabrication, spent-fuel processing, nuclear waste disposal, handling of radioisotopes, and environmental effects of these operations, are also treated. Table of Contents for this issue follows. GENERAL SAFETY CONSIDERATIONS: 1 The Safety Review and Approval Process for Space Nuclear Power Sources, G. L. Bennett; 19 Report on the American Nuclear Society International Topical Meeting: "The Safety, Status, and Future of Non-Commercial Reactors and Irradiation Facilities", E. G. Silver; 35 Book Review: Fission Product Transport Processes in Reactor Accidents, Proceedings to the International Centre tor Heat and Mass Processes T. S. Kress; 38 Fast Reactor Technology in the 1990s: A Summary of the 1990 International Fast Reactor Safety Meeting, A. E. Levin; ACCIDENT ANALYSIS: 56 Effects of Chemical Phenomena on LWR Severe Accident Fission Product Behavior, A. P. Malinauskas and T. S. Kress; CONTROL AND INSTRUMENTATION: 65 Technical Note: Safety Parameter Display Systems—10 Years Later, R. J. Eckenrode; 68 Potential Application of Neural Networks to the Operation of Nuclear Power Plants, R. E. Uhrig; DESIGN FEATURES: 80 Twenty-First DOE/NRC Nuclear Air-Cleaning Conference, R. R. Bellamy, D. W. Moeller, and M. W. First; 91 Impact of an Apparent Radiation Embrittlement Rate on the Life Expectancy of PWR Vessel Supports, R. D. Cheverton, G. C. Robinson, W. E. Pennell, and R. K. Nanstad; ENVIRONMENTAL EFFECTS: 103 Technical Note: The Impact of Offsite Factors on the Safety Performance of Small Nuclear Power Plants, Yu. D. Baranaev and A. N. Viktorov; WASTE AND SPENT FUEL MANAGEMENT: 109 Activities Related to Waste Management, Compiled by E. G. Silver; OPERATING EXPERIENCES: 118 Reactor Shutdown Experience, Compiled by J. W. Cletcher; 121 Selected Safety-Related Events, Compiled by G. A. Murphy; 123 Operating U.S. Power Reactors, Compiled by E. G. Silver; RECENT DEVELOPMENTS: 140 General Administrative Activities, Compiled by E. G. Silver; 150 Reports, Standards, and Safety Guides, D. S. Queener; 155 Status of Power-Reactor Licensing Activities, Compiled by E. G. Silver; 157 Proposed Rule Changes as of Sept. 30, 1990; ANNOUNCEMENTS: 79 MIT Offers Summer Program on Nuclear Power Reactor Safety; 108 Harvard School of Public Health Offers Several Short Courses; 139 Short Course and Workshop on Nuclear Criticality Safety at University of New Mexico; 160 SCK/CEN Announces Training Course on Emergency Planning and Response; 161 The Authors; 164 Indexes to Nuclear Safety, Volume 31

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Hydrologic response of artificially drained agricultural watersheds: insights from high-resolution integrated surface/subsurface simulations

Artificial drainage systems comprising subsurface networks of perforated pipes (tile drains) and engineered surface ditches are widely used to remove excess water from poorly drained agricultural regions. Artificial drainage lowers the water table by design but also has inadvertent effects on the watershed-scale hydrologic response with important implications for flood risk and nutrient exports. We investigated the effects of tile drains on watershed-scale hydrologic response in the Portage River, OH, Watershed using a high-resolution physically based integrated surface/subsurface hydrology model with recently developed capabilities to represent artificial drainage. Tile drains were found to enhance streamflow during times of low flow, generally consistent with previous studies. Streamflow flashiness was found to have a non-monotonic dependence on tile spacing with a minimum at intermediate spacings (∼50 m). Flashiness and the event hydrographs for small tile spacing were similar to the situation with no tiles, but flow paths from farm to stream were very different for those two end member cases, emphasizing the limitations of the stream hydrograph in characterizing hydrologic response. For typical tile spacings, peak flow can either be enhanced or attenuated by the presence of tiles, depending on the size of the event and the antecedent meteorological conditions. Tiles enhance peak flow when the event is below a threshold of ∼25 mm or when events arrive in dry conditions. Peak flow is reduced by tiles when events are large and arrive in conditions that are not overly dry. The dependence on event size and antecedent conditions is explained by differences in available storage and flow paths to the streams. These results provide additional insights into how tile drainage modulates event-scale hydrologic response, an important control on flood generation mechanisms and nutrient exports.

agricultural watershed↗
DOE OSTI2021

A Taxonomy for Classification and Comparison of Dataflows for GNN Accelerators

Recently, Graph Neural Networks (GNNs) have received a lot of interest because of their success in learning representations from graph structured data. However, GNNs exhibit different compute and memory characteristics compared to traditional Deep Neural Networks (DNNs). Graph convolutions require feature aggregations from neighboring nodes (known as the aggregation phase), which leads to highly irregular data accesses. GNNs also have a very regular compute phase that can be broken down to matrix multiplications (known as the combination phase). All recently proposed GNN accelerators utilize different dataflows and microarchitecture optimizations for these two phases. Different communication strategies between the two phases have been also used. However, as more custom GNN accelerators are proposed, the harder it is to qualitatively classify them and quantitatively contrast them. In this work, we present a taxonomy to describe several diverse dataflows for running GNN inference on accelerators. This provides a structured way to describe and compare the design-space of GNN accelerators.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

A geometric framework for momentum-based optimizers for low-rank training

Low-rank pre-training and fine-tuning have recently emerged as promising techniques for reducing the computational and storage costs of large neural networks. Training low-rank parameterizations typically relies on conventional optimizers such as heavy ball momentum methods or Adam. In this work, we identify and analyze potential difficulties that these training methods encounter when used to train low-rank parameterizations of weights. In particular, we show that classical momentum methods can struggle to converge to a local optimum due to the geometry of the underlying optimization landscape. To address this, we introduce novel training strategies derived from dynamical low-rank approximation, which explicitly account for the underlying geometric structure. Our approach leverages and combines tools from dynamical low-rank approximation and momentum-based optimization to design optimizers that respect the intrinsic geometry of the parameter space. We validate our methods through numerical experiments, demonstrating faster convergence, and stronger validation metrics at given parameter budgets.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗
DOE OSTI2026

NuGraph2 with context-aware inputs: physics-inspired improvements in semantic segmentation

Graph neural networks have recently shown strong promise for event reconstruction tasks in Liquid Argon Time Projection Chambers, yet their performance remains limited for underrepresented classes of particles, such as Michel electrons. In this work, we investigate physics-informed strategies to improve semantic segmentation within the NuGraph2 architecture. We explore three complementary approaches: (i) enriching the input representation with context-aware features derived from detector geometry and track continuity, (ii) introducing auxiliary decoders to capture class-level correlations, and (iii) incorporating energy-based regularization terms motivated by Michel electron energy distributions. Experiments on MicroBooNE public datasets show that physics-inspired feature augmentation yields the largest gains, particularly boosting Michel electron precision and recall by disentangling overlapping latent space regions. In contrast, auxiliary decoders and energy-regularization terms provided limited improvements, partly due to the hit-level nature of NuGraph2, which lacks explicit particle- or event-level representations. Our findings highlight that embedding physics context directly into node-level inputs is more effective than imposing task-specific auxiliary losses, and suggest that future hierarchical architectures such as NuGraph3, with explicit particle- and event-level reasoning, will provide a more natural setting for advanced decoders and physics-based regularization. The code for this work is publicly available on Github at https://github.com/vitorgrizzi/nugraph_phys/tree/main_phys.

Other Experiments↗
DOE OSTI2023

Merger identification through photometric bands, colours, and their errors

Aims. We present the application of a fully connected neural network (NN) for galaxy merger identification using exclusively photometric information. Our purpose is not only to test the method’s efficiency, but also to understand what merger properties the NN can learn and what their physical interpretation is. Methods. We created a class-balanced training dataset of 5860 galaxies split into mergers and non-mergers. The galaxy observations came from SDSS DR6 and were visually identified in Galaxy Zoo. The 2930 mergers were selected from known SDSS mergers and the respective non-mergers were the closest match in both redshift and r magnitude. The NN architecture was built by testing a different number of layers with different sizes and variations of the dropout rate. We compared input spaces constructed using: the five SDSS filters: u, g, r, i, and z; combinations of bands, colours, and their errors; six magnitude types; and variations of input normalization. Results. We find that the fibre magnitude errors contribute the most to the training accuracy. Studying the parameters from which they are calculated, we show that the input space built from the sky error background in the five SDSS bands alone leads to 92.64 ± 0.15% training accuracy. We also find that the input normalization, that is to say, how the data are presented to the NN, has a significant effect on the training performance. Conclusions. We conclude that, from all the SDSS photometric information, the sky error background is the most sensitive to merging processes. This finding is supported by an analysis of its five-band feature space by means of data visualization. Moreover, studying the plane of the g and r sky error bands shows that a decision boundary line is enough to achieve an accuracy of 91.59%.

79 ASTRONOMY AND ASTROPHYSICS↗
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 OSTI2024

A general mechanistic framework for cross-scale understanding of hot spots and hot moments in carbon and water fluxes

Semi-arid ecosystems, like those in the American Southwest, exert a massive impact on the interannual variability of carbon and water cycling. Unfortunately, these carbon and water fluxes are notoriously difficult to predict due to their high spatial and temporal variability, which is poorly captured by the current generation of vegetation models. Indeed, this region is exemplified by the ‘hot spots and hot moments’ concept, which states that small areas in space (‘hot spots’) and transient moments in time (‘hot moments’) exert an outsized influence on biogeochemical cycling. However, the factors that regulate these pulses in biogeochemical activity are unknown, as is their variability across space and time. These uncertainties severely limit efforts to better represent hot spots and hot moments in models. Here, we seek to develop a generalized method for detecting and quantifying the importance of hot spots and hot moments from individual plant to regional scales. Underpinning this method is our recently developed statistical approach for identifying hot spots and hot moments. By applying this method to semi-continuous measurements of plant water status, a depth profile of soil water potential, and ecosystem fluxes via eddy covariance, we will track the fate of water through the soil-plant-atmosphere continuum and identify the mechanistic drivers of these transient pulses in biogeochemical activity. Then, we will expand this approach across a broad network of Ameriflux towers, and apply a machine learning approach that will allow us to upscale measurements of hot spots and hot moments across the American Southwest and quantify their impact on carbon and water cycles. These products will allow us to identify hot spots and hot moments across spatio-temporal scales and will serve as crucial data sources for validating a new generation of models that can better capture highly dynamic carbon and water fluxes. The proposed method will be easily transferable across biomes and will serve as a framework for future research on hot spots and hot moments across the plant ecophysiology, biometeorology, and vegetation modeling communities.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Discovering Strong Gravitational Lenses in the Dark Energy Survey with Interactive Machine Learning and Crowd-sourced Inspection with Space Warps

We conduct a search for strong gravitational lenses in the Dark Energy Survey (DES) Year 6 imaging data. We implement a pre-trained Vision Transformer (ViT) for our machine learning (ML) architecture and adopt interactive machine learning to construct a training sample with multiple classes to address common types of false positives. Our ML model reduces ∼236 million DES cutout images to 22,564 targets of interest, including ∼85% of previously reported galaxy–galaxy lens candidates discovered in DES. These targets were visually inspected by citizen scientists, who ruled out ∼90% as false positives. Of the remaining 2618 candidates, 149 were expert-classified as “definite” lenses and 516 as “probable” lenses, for a total of 665 systems, with 147 of these candidates being newly identified. Additionally, we trained a second ViT to find double-source plane lens systems, finding at least one double-source system. Our main ViT excels at identifying galaxy–galaxy lenses, consistently assigning high scores to candidates with high expert assessments. The top 800 ViT-scored images include ∼100 of our “definite” lens candidates. This selection is an order of magnitude higher in purity than previous convolutional neural-network-based lens searches and demonstrates the feasibility of applying our methodology for discovering large samples of lenses in future surveys.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Seamless Hybrid-integrated Interconnect NEtwork (SHINE)

Large-scale date centers are becoming increasingly limited by the capacity of interconnects due to the rapid increasing power consumption and bandwidth demands driven by the rigorous down scaling of CMOS critical dimensions in according with the Moore’s Law. As a result of the continuing bandwidth scaling, optical interconnects based on parallel optics, where data are simultaneously transmitted along more than one lane, have becoming a preferred interconnected solution over copper cables. To building an efficient optical interconnect system, optical coupling between different elements, such as optical fibers, photonics chips, is a critical step, which is challenging due to the mode mismatch of varies optical ports. In this project, we proposed and demonstrated an optical interfacing scheme, which provided a light coupling approach with low energy loss, broadband operation, high alignment tolerance and compatible with scalable fabrication and testing. Here, the mode transformation is achieved by a total internal reflection from a chip integrated microscale freefrom surface. Due to the enormous light wavefront manipulating ability associated with the gigantic design space, the freeform couplers platform is universally applicable to a wide range of optical coupling scenarios, such as the interfacing between a waveguide to a fiber, another waveguide, surface normal device, or free space.

42 ENGINEERING↗
DOE OSTI2020

Global Prompt Proton Sensor Network: Monitoring Solar Energetic Protons Based on GPS Satellite Constellation

Energetic particle instruments on board Global Positioning System (GPS) satellites form a powerful global prompt proton sensor network (GPPSn) that provides an unprecedented opportunity to monitor and characterize solar energetic protons targeting the Earth. The medium-Earth orbits of the GPS constellation have the unique advantage of allowing solar energetic protons to be simultaneously measured from multiple points in both open- and closed-field line regions. Examining two example intervals of solar proton events, we showcase in this study how GPS proton data are prepared, calibrated, and utilized to reveal important features of solar protons, including their source, acceleration/scattering by interplanetary shocks, the relative position of Earth when impinged by these shocks, the shape of solar particle fronts, the access of solar protons inside the dynamic geomagnetic field, and temporally varying proton distributions in both energy and space. By comparing to Van Allen Probes data, GPS proton observations are further demonstrated not only to be useful for qualitatively monitoring the dynamics of solar protons but also for quantitative scientific research including determining cutoff L-shells. Finally, our results establish that this GPPSn can join forces with other existing solar proton monitors and contribute to observing, warning, understanding, and ultimately forecasting the incoming solar energetic proton events.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

Collective mode Brownian dynamics: A method for fast relaxation of statistical ensembles

Sampling equilibrium configurations of correlated systems of particles with long relaxation times (e.g., polymeric solutions) using conventional molecular dynamics and Monte Carlo methods can be challenging. This is especially true for systems with complicated, extended bond network topologies and other interactions that make the use and design of specialized relaxation protocols infeasible. We introduce a method based on Brownian dynamics simulations that can reduce the computational time it takes to reach equilibrium and draw decorrelated samples. Importantly, the method is completely agnostic to the particle configuration and the specifics of interparticle forces. In particular, we develop a mobility matrix that excites non-local, collective motion of N particles and can be computed efficiently in $\mathcal{O}(N)$ time. Particle motion in this scheme is computed by integrating the overdamped Langevin equation with an Euler–Maruyama scheme, in which Brownian displacements are drawn efficiently using a low-rank representation of the mobility matrix in position and wave space. We demonstrate the efficacy of the method with various examples from the realm of soft condensed matter and release a massively parallel implementation of the code as a plugin for the open-source package HOOMD-blue [J. A. Anderson et al., J. Comput. Phys. 227, 5342 (2008) and J. Glaser et al., Comput. Phys. Commun. 192, 97 (2015)] which runs on graphics processing units

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

FPGA-accelerated SpeckleNN with SNL for real-time X-ray single-particle imaging

We present the implementation of a specialized version of our previously published unified embedding model, SpeckleNN, for real-time speckle pattern classification in X-ray Single-Particle Imaging (SPI), using the SLAC Neural Network Library (SNL) on an FPGA platform. This hardware realization transitions SpeckleNN from a prototypic model into a practical edge solution, optimized for running inference near the detector in high-throughput X-ray free-electron laser (XFEL) facilities, such as those found at the Linac Coherent Light Source (LCLS). To address the resource constraints inherent in FPGAs, we developed a more specialized version of SpeckleNN. The original model, which was designed for broader classification across multiple biological samples, comprised ~5.6 million parameters. The new implementation, while reducing the parameter count to 64.6K (a 98.8% reduction), focuses on maintaining the model's essential functionality for real-time operation, achieving an accuracy of 90%. Furthermore, we compressed the latent space from 128 to 50 dimensions. This implementation was demonstrated on the KCU1500 FPGA board, utilizing 71% of available DSPs, 75% of LUTs, and 48% of FFs, with an average power consumption of 9.4W according to the Vivado post-implementation report. The FPGA performed inference on a single image with a latency of 45.015 microseconds at a 200 MHz clock rate. In comparison, running the same inference on an NVIDIA A100 GPU resulted in an average power consumption of ~73W and an image processing latency of around 400 microseconds. Our FPGA-accelerated version of SpeckleNN demonstrated significant improvements, achieving an 8.9 × speedup and a 7.8 × reduction in power consumption compared to the GPU implementation. Key advancements include model specialization and dynamic weight loading through SNL, which eliminates the need for time-consuming FPGA design re-synthesis, allowing fast and continuous deployment of models (re)trained online. These innovations enable real-time adaptive classification and efficient vetoing of speckle patterns, making SpeckleNN more suited for deployment in XFEL facilities. This implementation has the potential to significantly accelerate SPI experiments and enhance adaptability to evolving experimental conditions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Persistent Classification: Understanding Adversarial Attacks by Studying Decision Boundary Dynamics

ABSTRACT There are a number of hypotheses underlying the existence of adversarial examples for classification problems. These include the high‐dimensionality of the data, the high codimension in the ambient space of the data manifolds of interest, and that the structure of machine learning models may encourage classifiers to develop decision boundaries close to data points. This article proposes a new framework for studying adversarial examples that does not depend directly on the distance to the decision boundary. Similarly to the smoothed classifier literature, we define a (natural or adversarial) data point to be ( γ , σ)‐stable if the probability of the same classification is at least for points sampled in a Gaussian neighborhood of the point with a given standard deviation . We focus on studying the differences between persistence metrics along interpolants of natural and adversarial points. We show that adversarial examples have significantly lower persistence than natural examples for large neural networks in the context of the MNIST and ImageNet datasets. We connect this lack of persistence with decision boundary geometry by measuring angles of interpolants with respect to decision boundaries. Finally, we connect this approach with robustness by developing a manifold alignment gradient metric and demonstrating the increase in robustness that can be achieved when training with the addition of this metric.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗