Search NASA⌕ Search

SEARCH · Search NASA

Results for “data reconstruction”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

WRF-Chem & Data Assimilation: Reconstructing a Historic Dispersion Event

CONCLUSIONS • For this particular case and use, 3D-VAR was the most useful data assimilation method applied for modeling this plume. • Slight downside was the plume was modeled to move too quickly. • Whether this result is generalizable may be debatable based on the circumstances. • The atmospheric state altered how effective the data assimilation algorithms performed. • Nudging and 3D-VAR are not mutually exclusive, but their errors for wind speed may compound

Thomas, Andrew M. [Savannah River National Laborat↗
DOE OSTI2021

Detector Instrumentation, Energy Reconstruction and Data Analysis in the DUNE and NOvA Neutrino Experiments

Neutrino oscillations are so far the only experimental observation beyond the standard model since its development four decades ago. The remaining fundamental questions that can be answered by long-baseline neutrino oscillations are (1) the CP violation (whether neutrinos and antineutrinos behave the same way in oscillation), and (2) the mass hierarchy ($m_3>m_{1,2}$ or $m_{1,2}>m_3$), and (3) the octant of $\theta_{23}$ ($\theta_{23}>45^\circ, =45^\circ$ or $<45^\circ$). The PI's group at UCI focuses on long-baseline neutrino experiments DUNE and NOvA, key components of DOE's Intensity Frontier program. NOvA is the current major U.S.-based long-baseline neutrino experiment aiming to solve the mass hierarchy and $\theta_{23}$ octant. It is taking $\nu$ and $\bar{\nu}$ data from the NuMI beam at Fermilab. DUNE is the next-generation flagship neutrino experiment in the U.S. which is designed to decisively determine neutrino CP violation, mass hierarchy and $\theta_{23}$ octant. DUNE is Fermilab's first priority, and has been established as an international collaboration. During the period covered by the report, the PI's group has been exceptionally productive, with leading contributions to DUNE argon-purity monitoring detectors, NOvA/DUNE deep-learning reconstruction, and NOvA oscillation and near detector (ND) analyses. The group also is actively involved in DUNE/NOvA detector operation, data production, calibration, software upgrade and maintenance, and ProtoDUNE cold electronic QA/QC. Specifically, we have made accomplishments in the following four research projects: 1) Purity monitor R\&D and data analysis for DUNE, 2) Deep-learning-based energy reconstruction at NOvA and DUNE, 3) Oscillation analysis at NOvA, and 4) $\nu$-e elastic scattering analysis at NOvA.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Securing Federated Learning Against Active Reconstruction Attacks

Federated Learning (FL) has amassed notable attention for its ability to preserve user privacy while emphasizing the retainment of model training efficiency. Due to this potential, FL has been integrated in many domains, such as healthcare, finance, law, and industrial engineering, where data cannot be easily exchanged due to sensitive information and strict privacy laws. However, current research has indicated that FL protocols are easily compromised by active data reconstruction attacks employed by actively dishonest servers. The malicious modification of global model parameters allows an actively dishonest server to obtain a direct copy of users’ private data via gradient inversion. Here, this class of attacks is highly underexplored and continues to be a major challenge due to the intense threat model. In this paper, we propose OASIS as a scalable and modality-agnostic defense based on data augmentation that counteracts active data reconstruction attacks while preserving model performance. To generalize our defense, we uncover the intuition behind gradient inversion that enables these attacks and theoretically establish the conditions by which the defense can be considered robust regardless of attack design. From this, we formulate our defense with data augmentation that illustrates its ability to undermine the attack principle. We evaluate OASIS on five real-world datasets–two image-based (ImageNet and CIFAR100) and three text-based (Wikitext, Stack Overflow, and Shakespeare)–which span diverse uses cases such as vision tasks and language modeling. Comprehensive evaluations on these datasets exhibit the efficacy of OASIS and highlight its feasibility as a solution.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Integrated top-down process and voxel-based microstructure modeling for Ti-6Al-4V in laser wire direct energy deposition process

Laser-wire metal additive manufacturing (AM) is one of the ideal direct energy deposition (DED) processes for creating large-scale parts with a medium level of complexity. However, the DED process involves complex thermal signatures and wide length scales making the fabrication of realistic AM components and part qualification often reliant on experimental trial-and-error optimization. While experimental measurements over the full volume of a part are valuable and necessary, measuring the entire area of a part is significantly laborious and practically infeasible, particularly for large parts in terms of cost and rapid qualification. Therefore, in this work, we developed an effective thermal and microstructure modeling framework based on the Johnson–Mehl-Avrami-Kolmogorov (JMAK) and Koistinen & Marburger (KM) models through a top-down approach that considers plate distortion-affected thermal profiles. A voxel-by-voxel simulation method is used to predict individual phase fractions of Ti-6Al-4 V. The predicted results were validated through detailed metallurgical measurements. A combined voxel-by-voxel approach with a sparse data reconstruction technique produced a near-perfect reconstruction of the original data. This approach anticipates a significant reduction in data points and computation time and resources. Lastly, we conclude with potential extensions of this work to other modeling efforts.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Fast Hyperspectral Neutron Tomography

Hyperspectral neutron computed tomography is a tomographic imaging technique in which thousands of wavelength-specific neutron radiographs are measured for each tomographic view. In conventional hyperspectral reconstruction, data from each neutron wavelength bin are reconstructed separately, which is extremely time-consuming. These reconstructions often suffer from poor quality due to low signal-to-noise ratios. Consequently, material decomposition based on these reconstructions tends to produce inaccurate estimates of the material spectra and erroneous volumetric material separation. In this paper, we present two novel algorithms for processing hyperspectral neutron data: fast hyperspectral reconstruction and fast material decomposition. Both algorithms rely on a subspace decomposition procedure that transforms hyperspectral views into low-dimensional projection views within an intermediate subspace, where tomographic reconstruction is performed. The use of subspace decomposition dramatically reduces reconstruction time while reducing both noise and reconstruction artifacts. We apply our algorithms to both simulated and measured neutron data and demonstrate that they reduce computation and improve the quality of the results relative to conventional methods.

Chowdhury, Mohammad Samin Nur [Purdue University]↗
DOE OSTI2021

Using AI to build a hydrobiogeochemical soil model

Soil water content is a function of inputs from precipitation and outputs via evaporation, transpiration, lateral flow, and vertical percolation, and is sensitive to biogeochemical processes. As such, soils serve as an ideal integrator of atmospheric, hydrological, and biogeochemical processes affecting the water cycle. In addition, soil water retention capacity, infiltration rates, and hydraulic conductivity can buffer or exacerbate the effects of extreme precipitation events (e.g., flooding, runoff, subsurface transport, erosion, greenhouse gas emissions) and mitigate the impact of droughts and heat waves on land systems (e.g., fire, crop failure). However, integrating water cycle measurements spanning different land atmosphere compartments across scales is a fundamental barrier for numerical model predictability. A significant challenge is that each domain (soil, hydrology, biology, and atmosphere) typically collects different sets of data at different temporal and spatial frequencies/scales, and even different dimensionalities (2D vs 3D). To implement soil as an integrator of the water cycle in land models, we suggest that novel machine learning (ML) tools can be developed to effectively simulate complex landscapes across various domains and scales, extended to regions with sparse or no data. The ultimate goals are to improve predictive understanding of land-atmosphere interactions and to extend the predictability of current Earth System Models (ESMs) through better integration of hydrological and biogeochemical data. We envision a framework in which: (1) ML-aided data reconstructions enable the merger of data sources into a unified geospatial product; (2) automated detection techniques are used to improve the knowledge of complex soil processes and interactions; and (3) this knowledge is leveraged and incorporated into models through AI-based emulators to distinctly connect the land and atmospheric compartments of the water cycle in models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Optimizing Error-Bounded Lossy Compression for Scientific Data by Dynamic Spline Interpolation

Today's scientific simulations are producing vast volumes of data that cannot be stored and transferred efficiently because of limited storage capacity, parallel I/O bandwidth, and network bandwidth. The situation is getting worse over time because of the ever-increasing gap between relatively slow data transfer speed and fast-growing computation power in modern supercomputers. Error-bounded lossy compression is becoming one of the most critical techniques for resolving the big scientific data issue, in that it can significantly reduce the scientific data volume while guaranteeing that the reconstructed data is valid for users because of its compression-error-bounding feature. In this paper, we present a novel error-bounded lossy compressor based on a state-of-the-art prediction-based compression framework. Our solution exhibits substantially better compression quality than all of the existing error-bounded lossy compressors, with comparable compression speed. Specifically, our contribution is threefold. (1) We provide an in-depth analysis of why the best-existing prediction-based lossy compressor can only minimally improve the compression quality. (2) We propose a dynamic spline interpolation approach with a series of optimization strategies that can significantly improve the data prediction accuracy, substantially improving the compression quality in turn. (3) We perform a thorough evaluation using six real-world scientific simulation datasets across different science domains to evaluate our solution vs. all other related works. Experiments show that the compression ratio of our solution is higher than that of the second-best lossy compressor by 20%similar to 460% with the same error bound in most of the cases.

Zhao, Kai↗
DOE OSTI2025

A Survey on Error-Bounded Lossy Compression for Scientific Datasets

Error-bounded lossy compression has been effective in significantly reducing the data storage/transfer burden while preserving the reconstructed data fidelity very well. Many error-bounded lossy compressors have been developed for a wide range of parallel and distributed use cases for years. They are designed with distinct compression models and principles, such that each of them features particular pros and cons. In this article, we provide a comprehensive survey of emerging error-bounded lossy compression techniques. The key contribution is fourfold. (1) We summarize a novel taxonomy of lossy compression into six classic models. (2) We provide a comprehensive survey of 10 commonly used compression components/modules. (3) We summarized pros and cons of 47 state-of-the-art lossy compressors and present how state-of-the-art compressors are designed based on different compression techniques. (4) We discuss how customized compressors are designed for specific scientific applications and use-cases. We believe this survey is useful to multiple communities including scientific applications, high-performance computing, lossy compression, and big data.

Error-Bounded Lossy Compression↗
DOE OSTI2022

Enhanced detection of paramagnetic fluorine-19 magnetic resonance imaging agents using zero echo time sequence and compressed sensing

Fluorine-19 ( 19 F) magnetic resonance imaging (MRI) is an emerging technique offering specific detection of labeled cells in vivo. Lengthy acquisition times and modest signal-to-noise ratio (SNR) makes three-dimensional spin-density–weighted 19 F imaging challenging. Recent advances in tracer paramagnetic metallo-perfluorocarbon (MPFC) nanoemulsion probes have shown multifold SNR improvements due to an accelerated 19 F T 1 relaxation rate and a commensurate gain in imaging speed and averages. However, 19 F T 2 -reduction and increased linewidth limit the amount of metal additive in MPFC probes, thus constraining the ultimate SNR. To overcome these barriers, we describe a compressed sampling (CS) scheme, implemented using a “zero” echo time (ZTE) sequence, with data reconstructed via a sparsity-promoting algorithm. Our CS-ZTE scheme acquires k-space data using an undersampled spherical radial pattern and signal averaging. Image reconstruction employs off-the-shelf sparse solvers to solve a joint total variation and -norm regularized least square problem. To evaluate CS-ZTE, we performed simulations and acquired 19 F MRI data at 11.7 T in phantoms and mice receiving MPFC-labeled dendritic cells. For MPFC-labeled cells in vivo, we show SNR gains of ~6.3 × with 8-fold undersampling. We show that this enhancement is due to three mechanisms including undersampling and commensurate increase in signal averaging in a fixed scan time, denoising attributes from the CS algorithm, and paramagnetic reduction of T 1 . Importantly, 19 F image intensity analyses yield accurate estimates of absolute quantification of 19 F spins. Overall, the CS-ZTE method using MPFC probes achieves ultrafast imaging, a substantial boost in detection sensitivity, accurate 19 F spin quantification, and minimal image artifacts.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Denoising Autoencoder for Reconstructing Sensor Observation Data and Predicting Evapotranspiration: Noisy and Missing Values Repair and Uncertainty Quantification

Abstract Machine learning (ML) methods applied in scientific research often deal with interrelated features in high‐dimensional data. Reducing data noise and redundancy is needed to increase prediction accuracy and efficiency especially when dealing with data from field sensors. We explored an unsupervised learning method, the denoising autoencoder (DAE), to extract the underlying data structure from noisy raw data in the context of predicting hydrologic quantities from multiple field sensors. These sensors have intrinsic instrumental noise and occasional malfunctions that cause missing values. Our DAE neural network reconstructed meteorological sensor data containing noise and missing values to predict evapotranspiration in a mountainous watershed. The DAE reconstructed the sensor variables with a mean coefficient of determination value of 0.77 across 15 dimensions representing individual sensors. It reduced variance and bias uncertainties compared to a classical autoencoder model. The reconstruction quality varied across dimensions depending on their cross‐correlation and alignment with the underlying data structure. Uncertainties arising from the model structure were overall higher than those resulting from data corruption. We attached the DAE structure to a downstream ET‐prediction neural network in three formats and achieved reasonably accurate ET predictions . The use of the DAE notably reduced variance uncertainty in ET prediction. However, excessive variance reduction may be accompanied by an increase in bias due to the intrinsic bias‐variance tradeoff. Our method of evaluating and reducing uncertainties in aggregated data from different sources can be used to improve predictive models, process understanding, and uncertainty quantification for better water resource management. Plain Language Summary We present a machine learning method, namely the denoising autoencoder, which reduces the effects of data noise and missing values typically present in scientific data sets collected through sensor measurements. This method selects the most relevant information from noisy raw data collected by the instruments and fills in missing values. To demonstrate the effectiveness of our method, we applied it to predict evapotranspiration, a hydrologic variable that represents the water moved from the land surface to the atmosphere through a combination of evaporation and plant water use (transpiration). We also used a random sampling technique (the Monte Carlo method) to compare the uncertainty in the predictions when using the raw and noisy data versus the reconstructed data. The denoising process produced more accurate predictions of evapotranspiration with less uncertainty. Improved predictions of evapotranspiration can lead to a better understanding and accounting of water budgets. This ML approach is broadly suitable for a wide variety of applications that involve noisy sensor data with missing values. Key Points We used a denoising autoencoder (DAE) neural network to reduce noise in meteorological and soil sensor observations by on average We used Monte Carlo sampling to estimate the bias and variance of all model outputs, including uncertainty sources from data and the model We attached the DAE component to a downstream neural network to predict ET with the variance reduced by , compared to that without the DAE

denoising autoencoder↗
DOE OSTI2020

Planck 2018 results - VIII. Gravitational lensing

We present measurements of the cosmic microwave background (CMB) lensing potential using the final Planck 2018 temperature and polarization data. Using polarization maps filtered to account for the noise anisotropy, we increase the significance of the detection of lensing in the polarization maps from 5σ to 9σ. Combined with temperature, lensing is detected at 40σ. We present an extensive set of tests of the robustness of the lensing-potential power spectrum, and construct a minimum-variance estimator likelihood over lensing multipoles 8 ≤ L ≤ 400 (extending the range to lower L compared to 2015), which we use to constrain cosmological parameters. We find good consistency between lensing constraints and the results from the Planck CMB power spectra within the ΛCDM model. Combined with baryon density and other weak priors, the lensing analysis alone constrains σ 8 Ωm 0.25 = 0.589 ± 0.020 (1σ errors). Also combining with baryon acoustic oscillation data, we find tight individual parameter constraints, σ8 = 0.811 ± 0.019, H0 = 67.9-1.3+1.2 km s -1 Mpc -1 , and Ω m = 0.303 -0.018 +0.016 . Combining with Planck CMB power spectrum data, we measure σ 8 to better than 1% precision, finding σ 8 = 0.811 ± 0.006. CMB lensing reconstruction data are complementary to galaxy lensing data at lower redshift, having a different degeneracy direction in σ 8 - Ω m space; we find consistency with the lensing results from the Dark Energy Survey, and give combined lensing-only parameter constraints that are tighter than joint results using galaxy clustering. Using the Planck cosmic infrared background (CIB) maps as an additional tracer of high-redshift matter, we make a combined Planck-only estimate of the lensing potential over 60% of the sky with considerably more small-scale signal. We additionally demonstrate delensing of the Planck power spectra using the joint and individual lensing potential estimates, detecting a maximum removal of 40% of the lensing-induced power in all spectra. The improvement in the sharpening of the acoustic peaks by including both CIB and the quadratic lensing reconstruction is detected at high significance.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Get Non-Real: Randomized Sketching for High-Dimensional Non-Real Valued Data (Final Report)

In our final report for DE-C0022186, we describe the work we did on this grant towards the goals we proposed. Our first goal was characterizing fundamental limits for sketching of discrete high-dimensional matrices with low-dimensional structures. Our second main goal was designing algorithms for data reconstruction from sketches. We focus on approaches that are either specifically designed for non-real-valued data (binary, finite field) or that will translate more readily to that setting.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Biases in genome reconstruction from metagenomic data

Background Advances in sequencing, assembly, and assortment of contigs into species-specific bins has enabled the reconstruction of genomes from metagenomic data (MAGs). Though a powerful technique, it is difficult to determine whether assembly and binning techniques are accurate when applied to environmental metagenomes due to a lack of complete reference genome sequences against which to check the resulting MAGs. Methods We compared MAGs derived from an enrichment culture containing ~20 organisms to complete genome sequences of 10 organisms isolated from the enrichment culture. Factors commonly considered in binning software—nucleotide composition and sequence repetitiveness—were calculated for both the correctly binned and not-binned regions. This direct comparison revealed biases in sequence characteristics and gene content in the not-binned regions. Additionally, the composition of three public data sets representing MAGs reconstructed from the Tara Oceans metagenomic data was compared to a set of representative genomes available through NCBI RefSeq to verify that the biases identified were observable in more complex data sets and using three contemporary binning software packages. Results Repeat sequences were frequently not binned in the genome reconstruction processes, as were sequence regions with variant nucleotide composition. Genes encoded on the not-binned regions were strongly biased towards ribosomal RNAs, transfer RNAs, mobile element functions and genes of unknown function. Our results support genome reconstruction as a robust process and suggest that reconstructions determined to be >90% complete are likely to effectively represent organismal function; however, population-level genotypic heterogeneity in natural populations, such as uneven distribution of plasmids, can lead to incorrect inferences.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

TopFusion: Using Topological Feature Space for Fusion and Imputation in Multi-Modal Data

We present a novel multi-modal data fusion technique using topological features. The method, TopFusion, leverages the flexibility of topological data analysis tools (namely persistent homology and persistence images) to map multi-modal datasets into a common feature space by forming a new multi-channel persistence image. Each channel in the image is representative of a view of the data from a modality-dependent filtration. We demonstrate that the topological perspective we take allows for more effective data reconstruction, i.e. imputation. In particular, by performing imputation in topological feature space we are able to outperform the same imputation techniques applied to raw data or alternatively derived features. We show that TopFusion representations can be used as input to downstream deep learning-based computer vision models and doing so achieves comparable performance to other fusion methods for classification on two multi-modal datasets.

Myers, Audun D.↗
DOE OSTI2024

Repurposing of the Run 2 CMS High Level Trigger Infrastructure as a Cloud Resource for Offline Computing

The former CMS Run 2 High Level Trigger (HLT) farm is one of the largest contributors to CMS compute resources, providing about 25k job slots for offline computing. This CPU farm was initially employed as an opportunistic resource, exploited during inter-fill periods, in the LHC Run 2. Since then, it has become a nearly transparent extension of the CMS capacity at CERN, being located on-site at the LHC interaction point 5 (P5), where the CMS detector is installed. This resource has been configured to support the execution of critical CMS tasks, such as prompt detector data reconstruction. It can therefore be used in combination with the dedicated Tier 0 capacity at CERN, in order to process and absorb peaks in the stream of data coming from the CMS detector. The initial configuration for this resource, based on statically configured VMs, provided the required level of functionality. However, regular operations of this cluster revealed certain limitations compared to the resource provisioning and use model employed in the case of WLCG sites. A new configuration, based on a vacuum-like model, has been implemented for this resource in order to solve the detected shortcomings. This paper reports about this redeployment work on the permanent cloud for an enhanced support to CMS offline computing, comparing the former and new models’ respective functionalities, along with the commissioning effort for the new setup.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2026

Synthetic modeling of soft x-ray emissivity for magnetic island analysis in LTX-β

We present a synthetic soft x-ray (SXR) forward-modeling framework that characterizes the emissivity structure of rotating magnetic islands in the Lithium Tokamak eXperiment-β across space and time. Magnetic islands associated with tearing modes produce modulations in line-integrated SXR brightness. Traditional tomographic methods struggle to resolve these structures in devices with limited sightlines, resulting in an under-determined inversion problem, or otherwise necessitate equilibrium reconstruction data, preventing their use in active control. Our approach removes this challenge by combining a fully three-dimensional ray-tracing model with a time-dependent emissivity prescription of a $m/n$ = 2/1 magnetic island. The model is validated through helical island geometry, rotation measured by magnetic diagnostics, and equilibrium constraints from PSI-Tri reconstructions. Synthetic brightness signals are generated for each photodiode sightline from a modeled emissivity profile and directly compared with experimental data from a tangential midplane SXR array. By fitting the synthetic diagnostic output to the observed brightness across time, we simultaneously infer key geometric parameters-such as island radial location, island width, and rotation frequency-without requiring full tomographic reconstruction, all with an average absolute deviation of less than 3%. This work demonstrates that forward modeling with a single tangential array can extract key island parameters in a spherical tokamak, providing a computationally efficient alternative to conventional SXR tomography and providing a pathway toward real-time magnetic-island characterization in future devices.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Online randomized interpolative decomposition with a posteriori error estimator for temporal PDE data reduction

Traditional low-rank approximation is a powerful tool for compressing large data matrices that arise in simulations of partial differential equations (PDEs), but suffers from high computational cost and requires several passes over the PDE data. The compressed data may also lack interpretability thus making it difficult to identify feature patterns from the original data. Here, to address these issues, we present an online randomized algorithm to compute the interpolative decomposition (ID) of large-scale data matrices in situ. Compared to previous randomized IDs that used the QR decomposition to determine the column basis, we adopt a streaming ridge leverage score-based column subset selection algorithm that dynamically selects proper basis columns from the data and thus avoids an extra pass over the data to compute the coefficient matrix of the ID. In particular, we adopt a single-pass error estimator based on the non-adaptive Hutch++ algorithm to provide real-time error approximation for determining the best coefficients. As a result, our approach only needs a single pass over the original data and thus is suitable for large and high-dimensional matrices stored outside of core memory or generated in PDE simulations. A strategy to improve the accuracy of the reconstructed data gradient, when desired, within the ID framework is also presented. We provide numerical experiments on turbulent channel flow and ignition simulations, and on the NSTX Gas Puff Image dataset, comparing our algorithm with the offline ID algorithm to demonstrate its utility in real-world applications.

Column subset selection↗
DOE OSTI2022

NXREFINE

NXRefine implements a complete workflow for both data acquisition and reduction of single crystal diffuse x-ray scattering collected as a series of area detector frames during continuous sample rotation. The data are stored in HDF5 files, which conform to the NeXus data format standard. These files contain a comprehensive set of metadata, including the incident wavelength, powder calibration scans used to define detector distances, beam centers, and yaw, pitch, and roll corrections. Automated peak searches of the three-dimensional data arrays define a set of Bragg peaks that are used to define an orientation matrix, defined according to Busing and Levy, which is then used to transform the data from angular coordinates to reciprocal space coordinates using the CCTW program. Multiple rotations at different detector translations and offset rotation axes are combined to ensure that there are no gaps in the reconstructed data, and allow the construction of masks to eliminate detector artifacts due to Compton scattering within the sensor layers. Finally, the data can be transformed into 3D-ΔPDF maps using the punch-and-fill method after symmetrization.

OSBORN, RAYMOND↗