Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Relative accuracy”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

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

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

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

Reducing the Cost of CCSD Basis Set Extrapolation in Ab Initio Computational Thermochemistry

Here, a series of approximations to CCSD contributions in computational model chemistries is presented in the context of kcal mol –1 , kJ mol –1 , and 20 cm –1 theoretical predictions of total atomization energies, benchmarked within the HEAT+CH 4 test suite. A specific set of circumstances where MP2, without empirical scaling, may be used as an effective intermediate in the first two of these accuracy ranges was determined. However, SDQ-MP4, a method long used in pursuit of kcal mol –1 accuracy but relatively unstudied in the subchemical accuracy community, offers significant improvement over the quality of MP2 as a basis-set intermediate at significantly reduced cost compared to CCSD. Given this, we argue for SDQ-MP4 as the de facto CCSD basis-set intermediate in sub-chemical accuracy calculations when CCSD in a desired basis set becomes unaffordable. We additionally report on a “CBS-like” scheme, where MP2 and SDQ-MP4 are used in conjunction to create a “cheap” three-part approximation of large CCSD basis set limits. The data for the CCSD approximation schemes are organized in such a way that model chemistry developers can locate an analog of their current approach for the CCSD basis set limit and explore alternative intermediates that either decrease computational cost or increase computational accuracy. We also show, for a handful of molecules, that SDQ-MP4 shows promise as an effective basis-set intermediate for harmonic and fundamental frequency computations, allowing for zero-point corrections of nearly CCSD(T)/ANO1 quality using simple composite methods that only require CCSD(T)/ANO0.

Thorpe, James H. [Argonne National Laboratory (ANL↗
DOE OSTI2023

Microbiome-enabled genomic selection improves prediction accuracy for nitrogen-related traits in maize

Root-associated microbiomes in the rhizosphere (rhizobiomes) are increasingly known to play an important role in nutrient acquisition, stress tolerance, and disease resistance of plants. However, it remains largely unclear to what extent these rhizobiomes contribute to trait variation for different genotypes and if their inclusion in the genomic selection protocol can enhance prediction accuracy. To address these questions, we developed a microbiome-enabled genomic selection method that incorporated host SNPs and amplicon sequence variants from plant rhizobiomes in a maize diversity panel under high and low nitrogen (N) field conditions. Our cross-validation results showed that the microbiome-enabled genomic selection model significantly outperformed the conventional genomic selection model for nearly all time-series traits related to plant growth and N responses, with an average relative improvement of 3.7%. The improvement was more pronounced under low N conditions (8.4–40.2% of relative improvement), consistent with the view that some beneficial microbes can enhance N nutrient uptake, particularly in low N fields. However, our study could not definitively rule out the possibility that the observed improvement is partially due to the amplicon sequence variants being influenced by microenvironments. Using a high-dimensional mediation analysis method, our study has also identified microbial mediators that establish a link between plant genotype and phenotype. Some of the detected mediator microbes were previously reported to promote plant growth. The enhanced prediction accuracy of the microbiome-enabled genomic selection models, demonstrated in a single environment, serves as a proof-of-concept for the potential application of microbiome-enabled plant breeding for sustainable agriculture.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

Shake loss intensities in x-ray photoelectron spectroscopy: Theory, experiment, and atomic composition accuracy for MgO and related compounds

The relative intensities of XPS core levels, scaled by their photoionization cross sections, are regularly used to determine sample atomic composition. Cross sections, however, give the intensity to all possible final states for the core ionizations, not just to the main peak. This includes all intrinsic satellite structure (shake states and, for open shell systems, the different ionic multiplets). In practice, for solids, this is usually experimentally impossible to determine accurately because such a satellite structure sits on the inelastically scattered electron background and cannot be easily separated. Therefore, usually, only the intensity of the main peak is used. This limits the ultimate possible accuracy of XPS composition determination. The purpose of the present paper is to examine the contributions that a theoretical analysis of losses of intensity can make to improve quantitation. For an MgO single crystal, we show that the correct stoichiometry of 1:1 can be recovered using the theoretical analysis of the experimental MgO peak ratio intensities. For materials with a sufficient bandgap for the XPS main peaks to be separated from the scattered background, the intensity of main peaks can often be accurately determined. Thus, if one uses theory to calculate that fraction of the total intensity lost from a main peak into all its satellite structure, the intensity of just main peaks could then be used to more accurately determine relative atom % composition. This work tests this approach using a single crystal MgO (50% Mg, 50% O) standard. Ab initio electronic structure theory of representative MgO clusters is used to determine Hartree–Fock wave functions for the ground state and final ionized states corresponding to the main Mg 2p and O1s XPS peaks of the oxide. The sudden approximation, SA, is used to determine the fractional losses from these main peaks to shake satellites, which is found to be greater for O1s than Mg2p. This results in predicted “apparent composition” for stoichiometric MgO of 55.2% Mg, 44.8% O instead of the true 50% Mg, 50% O. Equivalent theory for CaO results in a predicted apparent Ca value of 53.4%. Experimentally, using Mg2s or 2p intensity ratio to O1s, we find values between 52.2% and 56.0% Mg using two crystals and four different instrument electron pass energies. The average value of the measurements is 54.5% Mg when corrected for the presence of an adventitious carbon overlayer and slight surface hydroxide. Though this agreement with theory may be somewhat fortuitous, given the potential experimental errors, which are fully discussed, it is similar to that in our earlier study on LiF. We also present preliminary experimental data on Mg(OH) 2 and MgSO 4 , which show a similar trend of apparently higher than 50% Mg, but we have no theory values. We are not yet able to experimentally test for validation of the difference between apparent composition for MgO (55.2% Mg) and CaO (53.4% Ca), owing to significant carbonate formation at the surface of the single crystal CaO. Here, an important conclusion is that the theoretical determination of shake losses, obtained with ab initio wavefunctions and the SA, is likely to be a useful way to calibrate the accuracy and reliability of compositions obtained from XPS intensities and merits further study.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Requirements on the gain calibration for LiteBIRD polarisation data with blind component separation

The detection of primordial B modes of the cosmic microwave background (CMB) could provide information about the early stages of the Universe's evolution. The faintness of this signal requires exquisite calibration accuracy and control of instrumental systematic effects which otherwise could bias the measurements. In this work, we study the impact of an imperfect relative polarisation gain calibration on the recovered value of the tensor-to-scalar ratio r for the LiteBIRD experiment, through the application of the blind Needlet Internal Linear Combination (NILC) foreground-cleaning method. We derive requirements on the relative calibration accuracy of the overall polarisation gain (Δg ν ) for each LiteBIRD frequency channel. Our results show that minimum variance techniques, as NILC, are less sensitive to systematic gain calibration uncertainties compared to a parametric approach, if the latter is not equipped with a proper modelling of these instrumental effects. In this study, the most stringent requirements are found in the channels where the CMB signal is relatively brighter, with the tightest constraints at 166 GHz (Δg ν ≈ 0.16%). This differs from the outcome of an analogous analysis performed with a parametric method, where the tightest requirements are obtained for the foreground-dominated channels. Gain calibration uncertainties, corresponding to the derived requirements, are then simultaneously propagated into all frequency channels. By doing so, we find that the overall impact on estimated r is lower than the total gain systematic budget for LiteBIRD approximately by a factor 5, due to the correlations of the impacts of gain calibration uncertainties in different frequency channels. In order to decouple the systematic effect from the specific choice of the model, we derive the requirements assuming constant spectral parameters for the foreground emission. To assess the robustness of the obtained results against more realistic scenarios, we repeat the analysis assuming sky models of intermediate and high complexity. In these further cases, we adopt an optimised NILC pipeline, called the Multi-Clustering NILC (MC-NILC). We find that the impact of gain calibration uncertainties on r is lower than the LiteBIRD gain systematics budget for the intermediate-complexity sky model. For the high-complexity case, instead, it would be necessary to tighten the requirements by a factor 1.8.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Semi-Analytic Functions to Calculate the Deposition Coefficients for Ice Crystal Vapor Growth in Bin and Bulk Microphysical Models

Numerical cloud models require estimates of the vapor growth rate for ice crystals. Current bulk and bin microphysical parameterizations generally assume that vapor growth is diffusion limited, though some parameterizations include the influence of surface attachment kinetics through a constant deposition coefficient. A parameterization for variable deposition coefficients is provided herein. The parameterization is an explicit function of the ambient ice supersaturation and temperature, and an implicit function of crystal dimensions and pressure. The parameterization is valid for variable surface types including growth by dislocations and growth by step nucleation. Deposition coefficients are predicted for the two primary growth directions of crystals, allowing for the evolution of the primary habits. Comparisons with benchmark calculations of instantaneous mass growth indicate that the parameterization is accurate to within a relative error of 1%. Parcel model simulations using Lagrangian microphysics as a benchmark indicate that the bulk parameterization captures the evolution of mass mixing ratio and fall speed with typical relative errors of less than 10%, whereas the average axis lengths can have errors of up to 20%. The bin model produces greater accuracy with relative errors often less that 10%. The deposition coefficient parameterization can be used in any bulk and bin scheme, with low error, if an equivalent volume spherical radius is provided.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Identifying Critical Infrastructure in Imagery Data Using Explainable Convolutional Neural Networks

To date, no method utilizing satellite imagery exists for detailing the locations and functions of critical infrastructure across the United States, making response to natural disasters and other events challenging due to complex infrastructural interdependencies. This paper presents a repeatable, transferable, and explainable method for critical infrastructure analysis and implementation of a robust model for critical infrastructure detection in satellite imagery. This model consists of a DenseNet-161 convolutional neural network, pretrained with the ImageNet database. The model was provided additional training with a custom dataset, containing nine infrastructure classes. The resultant analysis achieved an overall accuracy of 90%, with the highest accuracy for airports (97%), hydroelectric dams (96%), solar farms (94%), substations (91%), potable water tanks (93%), and hospitals (93%). Critical infrastructure types with relatively low accuracy are likely influenced by data commonality between similar infrastructure components for petroleum terminals (86%), water treatment plants (78%), and natural gas generation (78%). Local interpretable model-agnostic explanations (LIME) was integrated into the overall modeling pipeline to establish trust for users in critical infrastructure applications. The results demonstrate the effectiveness of a convolutional neural network approach for critical infrastructure identification, with higher than 90% accuracy in identifying six of the critical infrastructure facility types.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Towards Automated and High-Throughput Quantitative Sizing and Isotopic Analysis of Nanoparticles via Single Particle-ICP-TOF-MS

The work described herein assesses the ability to characterize gold nanoparticles (Au NPs) of 50 and 100 nm, as well as 60 nm silver shelled gold core nanospheres (Au/Ag NPs), for their mass, respective size, and isotopic composition in an automated and unattended fashion. Here, an innovative autosampler was employed to mix and transport the blanks, standards, and samples into a high-efficiency single particle (SP) introduction system for subsequent analysis by inductively coupled plasma–time of flight–mass spectrometry (ICP-TOF-MS). Optimized NP transport efficiency into the ICP-TOF-MS was determined to be >80%. This combination, SP-ICP-TOF-MS, allowed for high-throughput sample analysis. Specifically, 50 total samples (including blanks/standards) were analyzed over 8 h, to provide an accurate characterization of the NPs. This methodology was implemented over the course of 5 days to assess its long-term reproducibility. Impressively, the in-run and day-to-day variation of sample transport is assessed to be 3.54 and 9.52% relative standard deviation (%RSD), respectively. The determination of Au NP size and concentration was of <5% relative difference from the certified values over these time periods. Isotopic characterization of the 107 Ag/ 109 Ag particles (n = 132,630) over the course of the measurements was determined to be 1.0788 ± 0.0030 with high accuracy (0.23% relative difference) when compared to the multi-collector–ICP-MS determination.

36 MATERIALS SCIENCE↗
DOE OSTI2024

A large deformation multiphase continuum mechanics model for shock loading of soft porous materials

A large deformation, coupled finite-element (FE) model is developed to simulate the multiphase response of soft porous materials subjected to high strain-rate loading. The approach is based on the theory of porous media (TPM) at large deformations. Simplifications to the one-dimensional regime studied in the numerical simulations follow. An overview of several different time integration schemes is presented for the purpose of solving the nonlinear dynamic coupled balance of momenta (mixture and fluid) and balance of mass of the mixture equations. Numerical examples are presented for (i) verification against closed-form analytical solutions assuming small loads, (ii) demonstrating large deformation effects at high strain-rate, and (iii) showing differences in deformations between a single-phase elastodynamics model with occluded compressible pore fluid and a multiphase poroelastodynamics model at high strain-rate. The multiphase model shows that the relative motion of the pore fluid significantly dampens the deformation response of the solid skeleton as compared to the single-phase model, and makes it possible to extract quantitative values for the stresses of the different constituents, thereby allowing one to form preliminary conclusions about the onset of damage in the solid skeleton. The novelty of the current work is developing a multiphase, large deformation, mixture theory numerical model for high strain-rate loading of soft porous materials. It was discovered that explicit, adaptive time-stepping Runge–Kutta schemes offer high accuracy at relatively low cost when compared to traditional implicit or explicit central difference time-stepping schemes for shock-like loadings. Here, shock viscosity is added to the mixture momentum balance equation to regularize the shock front, and a stabilization term is added to the mixture mass balance equation to stabilize equal order interpolation finite elements for the coupled finite element solution of multiphase materials.

Engineering↗
DOE OSTI2022

Maintaining Trust in Reduction: Preserving the Accuracy of Quantities of Interest for Lossy Compression

As the growth of data sizes continues to outpace computational resources, there is a pressing need for data reduction techniques that can significantly reduce the amount of data and quantify the error incurred in compression. Compressing scientific data presents many challenges for reduction techniques since it is often on non-uniform or unstructured meshes, is from a high-dimensional space, and has many Quantities of Interests (QoIs) that need to be preserved. To illustrate these challenges, we focus on data from a large scale fusion code, XGC. XGC uses a Particle-In-Cell (PIC) technique which generates hundreds of PetaBytes (PBs) of data a day, from thousands of timesteps. XGC uses an unstructured mesh, and needs to compute many QoIs from the raw data, f.One critical aspect of the reduction is that we need to ensure that QoIs derived from the data (density, temperature, flux surface averaged momentums, etc.) maintain a relative high accuracy. We show that by compressing XGC data on the high-dimensional, nonuniform grid on which the data is defined, and adaptively quantizing the decomposed coefficients based on the characteristics of the QoIs, the compression ratios at various error tolerances obtained using a multilevel compressor (MGARD) increases more than ten times. We then present how to mathematically guarantee that the accuracy of the QoIs computed from the reduced f is preserved during the compression. We show that the error in the XGC density can be kept under a user-specified tolerance over 1000 timesteps of simulation using the mathematical QoI error control theory of MGARD, whereas traditional error control on the data to be reduced does not guarantee the accuracy of the QoIs.

Gong, Qian↗
DOE OSTI2022

Improving Predictive Capability in REHEDS Simulations with Fast, Accurate, and Consistent Non-Equilibrium Material Properties

Predictive design of REHEDS experiments with radiation-hydrodynamic simulations requires knowledge of material properties (e.g. equations of state (EOS), transport coefficients, and radiation physics). Interpreting experimental results requires accurate models of diagnostic observables (e.g. detailed emission, absorption, and scattering spectra). In conditions of Local Thermodynamic Equilibrium (LTE), these material properties and observables can be pre-computed with relatively high accuracy and subsequently tabulated on simple temperature-density grids for fast look-up by simulations. When radiation and electron temperatures fall out of equilibrium, however, non-LTE effects can profoundly change material properties and diagnostic signatures. Accurately and efficiently incorporating these non-LTE effects has been a longstanding challenge for simulations. At present, most simulations include non-LTE effects by invoking highly simplified inline models. These inline non-LTE models are both much slower than table look-up and significantly less accurate than the detailed models used to populate LTE tables and diagnose experimental data through post-processing or inversion. Because inline non-LTE models are slow, designers avoid them whenever possible, which leads to known inaccuracies from using tabular LTE. Because inline models are simple, they are inconsistent with tabular data from detailed models, leading to ill-known inaccuracies, and they cannot generate detailed synthetic diagnostics suitable for direct comparisons with experimental data. This project addresses the challenge of generating and utilizing efficient, accurate, and consistent non-equilibrium material data along three complementary but relatively independent research lines. First, we have developed a relatively fast and accurate non-LTE average-atom model based on density functional theory (DFT) that provides a complete set of EOS, transport, and radiative data, and have rigorously tested it against more sophisticated first-principles multi-atom DFT models, including time-dependent DFT. Next, we have developed a tabular scheme and interpolation methods that compactly capture non-LTE effects for use in simulations and have implemented these tables in the GORGON magneto-hydrodynamic (MHD) code. Finally, we have developed post-processing tools that use detailed tabulated non-LTE data to directly predict experimental observables from simulation output.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Regional Earthquake Ground Motion Simulations for Southern California With EQSIM: Insights From the 2008 Chino Hills, 2024 Highland Park, and 2021 Carson Earthquakes

This study presents physics-based, 3D simulations using the EQSIM framework for several earthquakes in the Los Angeles region. The primary objective was to assess the ability of deterministic physics-based ground motion simulations to reproduce the observed motions from historical events. The selected events included the mathematical equation M w 5.4 2008 Chino Hills, the mathematical equation M w 4.4 2024 Highland Park, and the mathematical equation M w 4.3 2021 Carson events. The simulated motions were evaluated by comparing the recorded and simulated seismograms, as well as the Fourier amplitude spectra, across multiple seismic stations. The SCEC 3D velocity model, CVM-S4.26.M01, was used to represent the regional geology, and ground motion simulations were carried out with a resolution of up to 5 Hz. The results indicate that the simulated motions captured the recorded motions up to approximately 4 Hz. While careful iterations regarding source parameters and corner frequencies were required, and, for the case of the Highland Park event, some of the near-source stations had relatively low accuracy, the present study established a positive step toward the utilization of physics-based simulations in practical applications. The computational efficiencies exhibited by EQSIM, especially on GPU clusters, further supported this assertion, as wall-clock times of simulations involving more than 10 billion grid points were as low as mathematical equation minutes. This permits ensemble simulations for a considered scenario event so that modeling uncertainties (e.g., source and geology) can be bracketed.

EQSIM↗
DOE OSTI2020

Real-time monitoring and prediction of water quality parameters and algae concentrations using microbial potentiometric sensor signals and machine learning tools

We report the overarching hypothesis of this study was that temporal microbial potentiometric sensor (MPS) signal patterns could be used to predict changes in commonly monitored water quality parameters by using artificial intelligence/machine learning tools. To test this hypothesis, the study first examines a proof of concept by correlating between MPS's signals and high algae concentrations in an algal cultivation pond. Then, the study expanded upon these findings and examined if multiple water quality parameters could be predicted in real surface waters, like irrigation canals. Signals generated between the MPS sensors and other water quality sensors maintained by an Arizona utility company, including algae and chlorophyll, were collected in real time at time intervals of 30 min over a period of 9 months. Data from the MPS system and data collected by the utility company were used to train the ML/AI algorithms and compare the predicted with actual water quality parameters and algae concentrations. Based on the composite signal obtained from the MPS, the ML/AI was used to predict the canal surface water's turbidity, conductivity, chlorophyll, and blue-green algae (BGA), dissolved oxygen (DO), and pH, and predicted values were compared to the measured values. Initial testing in the algal cultivation pond revealed a strong linear correlation (R 2 = 0.87) between mixed liquor suspended solids (MLSS) and the MPSs' composite signals. The Normalized Root Mean Square Error (NRMSE) between the predicted values and measured values were <6.5%, except for the DO, which was 10.45%. The results demonstrate the usefulness of MPSs to predict key surface water quality parameters through a single composite signal, when the ML/AI tools are used conjunctively to disaggregate these signal components. The maintenance-free MPS offers a novel and cost-effective approach to monitor numerous water quality parameters at once with relatively high accuracy.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Cryo-EM model validation recommendations based on outcomes of the 2019 EMDataResource challenge

This paper describes outcomes of the 2019 Cryo-EM Model Challenge. The goals were to (1) assess the quality of models that can be produced from cryogenic electron microscopy (cryo-EM) maps using current modeling software, (2) evaluate reproducibility of modeling results from different software developers and users and (3) compare performance of current metrics used for model evaluation, particularly Fit-to-Map metrics, with focus on near-atomic resolution. Our findings demonstrate the relatively high accuracy and reproducibility of cryo-EM models derived by 13 participating teams from four benchmark maps, including three forming a resolution series (1.8 to 3.1 Å). The results permit specific recommendations to be made about validating near-atomic cryo-EM structures both in the context of individual experiments and structure data archives such as the Protein Data Bank. We recommend the adoption of multiple scoring parameters to provide full and objective annotation and assessment of the model, reflective of the observed cryo-EM map density.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Direct and rapid uranium isotopic analysis of environmental sample swipes via microextraction – LS-APGD/Orbitrap mass spectrometry

Accurate and precise isotopic analysis of actinides collected on environmental sample swipes from within nuclear facilities is an important safeguarding measure for detecting undeclared materials and activities. Traditional isotope ratio (IR) analysis of actinides using a bulk digestion approach can be laborious and time-consuming. Recently, direct analysis of environmental swipe samples using a microextraction approach has been explored as an alternative to conventional bulk digestion methods. The present study further evaluates this approach for accurate and precise isotopic analysis of uranium (U) in cotton swipe samples using the liquid sampling-atmospheric pressure glow discharge (LS-APGD)/Orbitrap-FTMS Booster detection system. The instrumental parameters were optimized, and quantitative capabilities were demonstrated through excellent linearity (R2 = 0.99) across a deposited mass range of 1 to 500 ng, with a limit of detection of 90 pg (in a 2 × 4 mm region) for total U; well below typical U concentrations (ng–mg range) in environmental sample swipes. The 235U/238U ratio showed excellent accuracy (−0.3% relative difference from the certificate value), with ∼20× improvement over previous microextraction-Orbitrap methods. Precision (∼1% relative standard deviation) was also greatly enhanced by ∼5×. This study also demonstrates that figures of merit achieved with neat U solution were not highly degraded when various external interfering elements/matrices were introduced. Finally, the developed method was successfully applied to analyze practical swipe samples representative of laboratory and outdoor environments. The presented figures of merit and applicability to practical samples further validate the capability of the platform for rapid analysis (<7 minutes) of environmental swipe samples.

Shrestha, Suraj [Clemson University, SC]↗
DOE OSTI2025

Development of a data-driven neural network model for electron thermal transport in NSTX

A data-driven electron thermal transport neural network (ETT-NN) model, trained on TRANSP interpretative analysis results of National Spherical Torus Experiment (NSTX), was developed to enable faster and more accurate ETT computation for spherical tokamaks (STs). The model incorporates both convolutional NNs and recurrent NNs, allowing it to simultaneously account for the spatial and temporal non-localities and multi-scale features of turbulent transport, which have been considered only in a limited manner in conventional models. The model was validated through interpretative analysis and predictive simulations using Tokamak Reactor Integrated Automated Suite for Simulation and Computation, demonstrating relatively high accuracy. Additionally, parameter scans were performed on test discharges known to exhibit specific turbulent modes, such as microtearing mode, trapped electron mode, kinetic ballooning mode, and electron temperature gradient mode. The scanning results revealed that the ETT-NN model exhibits the same trends as those observed in conventional gyrokinetic simulations or theories, while also capturing the global nature of turbulent transport, indicating that the data-driven model accurately reflects the underlying physical characteristics. Furthermore, due to the dimensionless nature of the model, we can feasibly expand its applicability by incorporating data from other devices and uncovering the characteristics of ETT in STs in the future.

NSTX↗
DOE OSTI2024

Exploiting a derivative discontinuity estimate for accurate G0W0 ionization potentials and electron affinities

Abstract The GW approximation has become an important tool for predicting charged excitations of isolated molecules and condensed systems. Its popularity can be attributed to many factors, including a favorable scaling and relatively good accuracy. In practical applications, the GW is often performed as a one-shot perturbation known as G 0 W 0 . Unfortunately, G 0 W 0 suffers from a strong starting point dependence and is often not as accurate as one would need. Self-consistent GW methodologies alleviate these problems but come with a marked increase in computational cost. In this manuscript, we propose the use of an estimate of the exchange-correlation derivative discontinuity to provide a remarkably good starting point for G 0 W 0 calculations, yielding ionization potentials and electron affinities with eigenvalue self-consistent GW quality at no additional cost. We assess the quality of the resulting methodology with the GW 100 benchmark set and compare its advantages over other similar methods.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

ASK: Adversarial Soft k-Nearest Neighbor Attack and Defense

K-Nearest Neighbor (kNN)-based deep learning methods have been applied to many applications due to their simplicity and geometric interpretability. However, the robustness of kNN-based deep classification models has not been thoroughly explored and kNN attack strategies are underdeveloped. In this paper, we first propose an Adversarial Soft kNN (ASK) loss for developing more effective kNN-based deep neural network attack strategies and designing better defense methods against them. Our ASK loss provides a differentiable surrogate of the expected kNN classification error. It is also interpretable as it preserves the mutual information between the perturbed input and the in-class-reference data. We use the ASK loss to design a novel attack method called the ASK-Attack (ASK-Atk), which shows superior attack efficiency and accuracy degradation relative to previous kNN attacks on hidden layers. We then derive an ASK-Defense (ASK-Def) method that optimizes the worst-case ASK training loss. Experiments on CIFAR-10 (ImageNet) show that (i) ASK-Atk achieves ≥13% (≥ 13% ) improvement in attack success rate over previous kNN attacks, and (ii) ASK-Def outperforms the conventional adversarial training method by ≥ 6.9% (≥ 3.5% ) in terms of robustness improvement. Relevant codes are available at https://github.com/wangren09/ASK .

97 MATHEMATICS AND COMPUTING↗