Nonnegative matrix factorization-based blind source separation for full-field and high-resolution modal identification from video
Not Available
Silva, Moisés (ORCID:0000000178973978)↗
SEARCH · Search NASA
Results for “separation factor”
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.
Exploring QCD Factorization at Moderate Energy Scales
Asymptotic freedom in QCD facilitates the use of partonic degrees of freedom over short distances, but physical processes are sensitive to a wide range of scales. Thus, it is necessary in QCD calculations to utilize a factorization scheme to separate a process into perturbative and non-perturbative factors. This separation relies on an assumption that one energy scale is infinitely larger than the other scales involved in the process. However, much experimental research in areas such as nucleon structure and quark-hadron duality occur at more moderate energy scales where that basic assumption may not be true but perturbative calculations should still be useful. Thus, an exploration of the limits of factorization at more moderate energy scales is needed. This dissertation examines various aspects of factorization at these energy scales first by applying the necessary approximations to a simple model where exact calculations are possible and so the effects of these approximations can be quantified. This is followed by examining areas where corrections are known to be needed. First is an exploration of target mass corrections (TMCs) in the case of deep inelastic scattering (DIS), including a discussion of what large corrections imply about the target structure. Second, is a general examination of PDFs and FFs at moderate scales. Third, I will discuss how this fits into the long-term effort to study the transition between small transverse momentum (generated non-perturbatively) and large transverse momentum (generated in the hard process)
Moffat, Eric↗
Anomaly detection in collider physics via factorized observables
To maximize the discovery potential of high-energy colliders, experimental searches should be sensitive to unforeseen new physics scenarios. This goal has motivated the use of machine learning for unsupervised anomaly detection. In this paper, we introduce a new anomaly detection strategy called : factorized observables for regressing conditional expectations. Our approach is based on the inductive bias of factorization, which is the idea that the physics governing different energy scales can be treated as approximately independent. Assuming factorization holds separately for signal and background processes, the appearance of nontrivial correlations between low- and high-energy observables is a robust indicator of new physics. Under the most restrictive form of factorization, a machine-learned model trained to identify such correlations will in fact converge to the optimal new physics classifier. We test on a benchmark anomaly detection task for the Large Hadron Collider involving collimated sprays of particles called jets. By teasing out correlations between the kinematics and substructure of jets, our method can reliably extract percent-level signal fractions. This strategy for uncovering new physics adds to the growing toolbox of anomaly detection methods for collider physics with a complementary set of assumptions. Published by the American Physical Society 2024
Astronomy & Astrophysics↗
Physics and chemistry from parsimonious representations: image analysis via invariant variational autoencoders
Electron, optical, and scanning probe microscopy methods are generating ever increasing volume of image data containing information on atomic and mesoscale structures and functionalities. This necessitates the development of the machine learning methods for discovery of physical and chemical phenomena from the data, such as manifestations of symmetry breaking phenomena in electron and scanning tunneling microscopy images, or variability of the nanoparticles. Variational autoencoders (VAEs) are emerging as a powerful paradigm for the unsupervised data analysis, allowing to disentangle the factors of variability and discover optimal parsimonious representation. Here, we summarize recent developments in VAEs, covering the basic principles and intuition behind the VAEs. The invariant VAEs are introduced as an approach to accommodate scale and translation invariances present in imaging data and separate known factors of variations from the ones to be discovered. We further describe the opportunities enabled by the control over VAE architecture, including conditional, semi-supervised, and joint VAEs. Several case studies of VAE applications for toy models and experimental datasets in Scanning Transmission Electron Microscopy are discussed, emphasizing the deep connection between VAE and basic physical principles. Python codes and datasets discussed in this article are available at https://github.com/saimani5/VAE-tutorials and can be used by researchers as an application guide when applying these to their own datasets.
36 MATERIALS SCIENCE↗
Quasielastic lepton-nucleus scattering and the correlated Fermi gas model
The neutrino research program in the coming decades will require improved precision. A major source of uncertainty is the interaction of neutrinos with nuclei that serve as targets for such experiments. Broadly speaking, this interaction often depends, e.g., for charge-current quasielastic scattering, on the combination of “nucleon physics,” expressed by form factors, and “nuclear physics,” expressed by a nuclear model. It is important to get a good handle on both. We present a fully analytic implementation of the correlated Fermi gas model for electron-nucleus and charge-current quasielastic neutrino-nucleus scattering. The implementation is used to compare separately form factors and nuclear model effects for both electron-carbon and neutrino-carbon scattering data. Published by the American Physical Society 2025
Bhattacharya, Bhubanjyoti (ORCID:000000032238321X)↗
Anomalous dimensions from soft Regge constants
Using an effective field theory (EFT) formalism for forward scattering, we reconsider the factorization of 2 → 2 scattering amplitudes in the Regge limit. Expanding the amplitude in gauge invariant operators labelled by the number of Glauber exchanges, allows us to further factorize the standard impact factors into separate collinear and soft functions. The soft functions are universal, and describe radiative corrections to the Reggeized gluon states exchanged by the collinear projectiles. Remarkably, we find that the one-loop soft function for the single Reggeized gluon state is given to O(ϵ) in terms of the two-loop cusp and two-loop rapidity anomalous dimensions. We argue that this iterative structure follows from the simple action of crossing symmetry in the forward scattering limit, which in the EFT allows us to replace the divergent part of a soft loop by a much simpler Glauber loop. We use this correspondence to provide a simple calculation of the two-loop Regge trajectory using the EFT. We then explore its implications at higher perturbative orders, and derive the maximally matter dependent contributions to the Regge trajectory to all loop orders, i.e. the terms ~ α$^{k+1}_{s}$n$^{k}_{f}$ for any k, where n f is the number of massless flavors. These simplifications suggests that the EFT approach to the Regge limit will be helpful to explore and further understand the structure of the Regge limit.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Efficient use of quantum computers for collider physics
Most observables at particle colliders involve physics at a wide variety of distance scales. Due to asymptotic freedom of the strong interaction, the physics at short distances can be calculated reliably using perturbative techniques, while long distance physics is non-perturbative in nature. Factorization theorems separate the contributions from different scales, allowing to identify the pieces that can be determined perturbatively from those that require non-perturbative information, and if the non-perturbative pieces can be reliably determined, one can use experimental measurements to extract the short distance effects, sensitive to possible new physics. Without the ability to compute the non-perturbative ingredients from first principles one typically identifies observables for which the non-perturbative information is universal in the sense that it can be extracted from some experimental observables and then used to predict other observables. In this paper we argue that the future ability to use quantum computers to calculate non-perturbative matrix elements from first principles will allow to make predictions for observables with non-universal non-perturbative long-distance physics.
Algorithms and Theoretical Developments↗
Validation of 3D cryoEM single-particle reconstruction correctness and handedness with Ewald's sphere correction
The correct description of quantum scattering places the observed scattering contributions on the Ewald's sphere and its Friedel mate copy. In electron microscopy, due to the large radius of the Ewald's sphere, these scattering contributions are typically merged during data analysis. We present an approach that separates and factorizes those contributions into real and imaginary components of the image. When an inverted solution is calculated, the map derived from the real component of the image generates an inverted solution, while the map derived from the imaginary component of the image generates an inverted and sign-flipped solution. Therefore, the sign of correlation between reconstructions derived from the real and imaginary components provides the automatic determination of handedness and additional validation for the quality of 3D reconstructions. The factorization and its implementation are robust enough to be routinely used in single-particle reconstructions, even at resolutions below the limit where the curvature of the Ewald's sphere affects the overall signal-to-noise ratio.
Chemistry↗
Multiwavelength anomalous X-ray diffraction for combined imaging of atomic displacement and strain
The X-ray Bragg coherent diffractive imaging (CDI) technique assumes that the structure factor holds constant over the measured crystal. This approximation breaks down for materials exhibiting variations in the unit-cell configuration, such as piezo- and ferroelectrics. In that case, the strain field cannot be reliably determined from the reconstruction because the lattice deformation and the structure factor contribute concomitantly. Proposed here is a solution to this problem achieved by combining Bragg CDI and the multiwavelength anomalous diffraction approach that measures a Friedel pair of reflections at two different photon energies near an absorption edge. Comparing the obtained reconstructions with a parametric model that includes calculating the scattering amplitude as a function of wavelength and the unit-cell configuration, the contributions of the lattice deformation and the structure factor are separated. Simulations of the ferroelectric material BaTiO 3 demonstrate the possibility of simultaneous probing of the strain and displacement of the Ti atoms. The proposed method opens up an opportunity to apply coherent X-ray diffraction for nanoscale-resolved 3D mapping of polarization domains in micro- and nanocrystals.
lattice deformation↗
Factors that Impact Extraction and Separation : Guiding Development for Materials to Sequester PFAS
Explore the source record for details and available documents.
Knight, Andrew William↗
Coffea -- Columnar Object Framework For Effective Analysis
The coffea framework provides a new approach to High-Energy Physics analysis, via columnar operations, that improves time-to-insight, scalability, portability, and reproducibility of analysis. It is implemented with the Python programming language, the scientific python package ecosystem, and commodity big data technologies. To achieve this suite of improvements across many use cases, coffea takes a factorized approach, separating the analysis implementation and data delivery scheme. All analysis operations are implemented using the NumPy or awkward-array packages which are wrapped to yield user code whose purpose is quickly intuited. Various data delivery schemes are wrapped into a common front-end which accepts user inputs and code, and returns user defined outputs. We will discuss our experience in implementing analysis of CMS data using the coffea framework along with a discussion of the user experience and future directions.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Physical discovery in representation learning via conditioning on prior knowledge
Recent advances in electron, scanning probe, optical, and chemical imaging and spectroscopy yield bespoke data sets containing the information of structure and functionality of complex systems. In many cases, the resulting data sets are underpinned by low-dimensional simple representations encoding the factors of variability within the data. The representation learning methods seek to discover these factors of variability, ideally further connecting them with relevant physical mechanisms. However, generally, the task of identifying the latent variables corresponding to actual physical mechanisms is extremely complex. Here, we present an empirical study of an approach based on conditioning the data on the known (continuous) physical parameters and systematically compare it with the previously introduced approach based on the invariant variational autoencoders. The conditional variational autoencoder (cVAE) approach does not rely on the existence of the invariant transforms and hence allows for much greater flexibility and applicability. Interestingly, cVAE allows for limited extrapolation outside of the original domain of the conditional variable. However, this extrapolation is limited compared to the cases when true physical mechanisms are known, and the physical factor of variability can be disentangled in full. We further show that introducing the known conditioning results in the simplification of the latent distribution if the conditioning vector is correlated with the factor of variability in the data, thus allowing us to separate relevant physical factors. We initially demonstrate this approach using 1D and 2D examples on a synthetic data set and then extend it to the analysis of experimental data on ferroelectric domain dynamics visualized via piezoresponse force microscopy.
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
Molecular Simulations Probing the Adsorption and Diffusion of Ammonia, Nitrogen, Hydrogen, and Their Mixtures in Bulk MFI Zeolite and MFI Nanosheets at High Temperature and Pressure
Recent advances in the synthesis of MFI zeolite nanosheets have led to highly selective membranes that are promising candidates for small-scale ammonia separation from ammonia/nitrogen/hydrogen mixtures in distributed green ammonia production plants. Using force-field-based molecular simulations, we evaluate the performance of bulk all-silica MFI zeolite and a 3 nm thick all-silica MFI nanosheet for ammonia/nitrogen/hydrogen separations at moderate temperature and pressure conditions (T = 373 K, p = 5 bar) as well as conditions relevant for a membrane-based reactor–separator process (T = 523 or 623 K, p = 80 bar). Isobaric–isothermal Gibbs ensemble Monte Carlo (GEMC) simulations were carried out to understand the selective adsorption behavior for these mixtures. Molecular dynamics simulations in the canonical ensemble were performed to examine the transport behavior of ammonia in a mixture with nitrogen or hydrogen at loadings obtained from the GEMC simulations. Our results show that both bulk MFI and the nanosheet with explicit surface silanols are highly selective toward ammonia adsorption, but the adsorption selectivity decreases by factors of 4 and 10 for ammonia/nitrogen and ammonia/hydrogen mixtures as the temperature is increased from 373 to 623 K. Conversely, the diffusion selectivities toward ammonia are more favorable at process-relevant temperatures and for the MFI nanosheet. Here, at T = 523 K and p = 80 bar, our simulations predict overall separation factors of 3.8 ± 0.2 and 4.5 ± 0.3 for ammonia/nitrogen and ammonia/hydrogen mixtures, respectively, in the bulk MFI zeolite, and separation factors of 8.6 ± 0.4 and 12.5 ± 0.8, respectively, for the MFI nanosheet.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Quantum entropy as a harbinger of factorizability
Deeply inelastic scattering (DIS) is a powerful probe for investigating the QCD structure of hadronic matter and testing the standard model (SM). DIS can be described through QCD factorization theorems which separate contributions to the scattering interaction arising from disparate scales — e.g ., with nonperturbative matrix elements associated with long distances and a perturbative hard scattering kernel applying to short-distance parton-level interactions. The fundamental underpinnings of factorization may be recast in the quantum-theoretic terms of entanglement, (de)coherence, and system localization in a fashion which sheds complementary light on the dynamics at work in DIS from QCD bound states. In this Letter, we propose and quantitatively test such a quantum-information theoretic approach for dissecting factorization in DIS and its domain of validity; we employ metrics associated with quantum entanglement such as a differential quantum entropy and associated Kullback-Leibler (KL) divergences in numerical tests. We deploy these methods on an archetypal quark-spectator model of the proton, for which we monitor quantum decoherence in DIS as underlying model parameters are varied. On this basis, we demonstrate quantitatively how factorization-breaking effects may be imprinted on quantum entropies in a kinematic regime where leading-twist factorization increasingly receives large corrections from finite- Q 2 effects; our findings suggest potential applications of quantum simulation to QCD systems and their interactions.
Deep inelastic scattering↗
Microbe-Encapsulated Silica Gel Biosorbents for Selective Extraction of Scandium from Coal Byproducts
Establishing an inexpensive and environmentally friendly scandium (Sc) supply is critical for the development of clean energy technologies. However, refining Sc from Sc-bearing sources using current technologies poses economic and environmental challenges due to its low concentration relative to base metals and chemically similar lanthanides. In this work, we developed a biosorption-based flow-through process for extraction of Sc from low-grade feedstocks. A novel biosorbent material was synthesized by encapsulating Arthrobacter nicotianae, a bacterium that selectively adsorbs Sc, within a porous silica matrix. Through the use of several complementary microscopy techniques, we demonstrated a highly porous biosorbent structure with a high cell-loading density and a homogenous cell distribution. Batch adsorption assays revealed selective Sc adsorption over lanthanides and common based metals, with the exception of Fe(III), as evidenced by separation factors greater than 50. The MESG particles were packed into fixed-bed columns to enable adsorption under flow-through conditions, which demonstrated effective Sc extraction at flow rates up to 0.08 cm/s and high stability for reuse; greater than 95% of the adsorption capacity was maintained after 10 consecutive adsorption/desorption cycles. When applied to a lignite coal leachate, whose Fe content was depleted through pH-mediated precipitation, the MESG particles yielded Sc breakthrough at 30 bed volumes, whereas all other metals broke through after only a few bed volumes. Following desorption, an eluate with a 124-fold increase in Sc purity was achieved relative to the lignite leachate with Sc constituting 96.4% of the total REEs. This study established a rapid, facile, and scalable cell immobilization approach that enables the use of microbial biomass for Sc recovery from low-grade sources.
Dong, Ziye↗
Estimating region-specific fuel economy in the United States from real-world driving cycles
Here we describe a method for estimating region-specific real-world light-duty vehicle fuel economy in the United States that is unique in both the size and representativeness of real-world driving that was considered, and for its ability to model regional variations in driving patterns. Over one million miles of national driving data were used to select real-world cycles representative of observed trip categories. The six cycles were compared to U.S. legislative cycles, revealing some key differences. Finally, a set of cycle weighting factors for 533 separate U.S. regions was derived from annual traffic statistics. Applying this method, it was found that regional fuel economy varies due to differences in driving patterns alone and that rural driving patterns lead to improved fuel economy (for conventional vehicles). The driving cycles and regional weighting factors described here are useful for testing and simulation studies, specifically those sensitive to regional variations in driving patterns.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
Direct quarkonium production in DIS from a joint CGC and NRQCD framework
We compute the differential cross section for direct quarkonium production in high-energy electron-nucleus collisions at small 𝑥. Our computation is performed within the nonrelativistic QCD factorization formalism that separates the calculation into short distance coefficients and long distance matrix elements that depend on the color and spin of the state. We obtain the short distance coefficients of the production of the heavy quark pair within the framework of the color glass condensate effective field theory, which resums coherent multiple interactions of the heavy quark pair with the nucleus to all orders. Our results are expressed as the convolution of perturbatively calculable functions with multipoint lightlike Wilson line correlators. In the correlation limit, we establish the correspondence between our color glass condensate formulation with calculations employing the transverse momentum dependent (TMD) framework. We extend this correspondence by resumming kinematic power corrections within the improved TMD framework, which interpolates between the TMD formalism and 𝑘 ⊥ -factorization formalism. We present a detailed numerical analysis, focusing on 𝐽/𝜓 production in the kinematics accessible at the future Electron-Ion Collider, highlighting the importance of genuine higher-order saturation contributions when the electron collides with a large nucleus. Our results are also valid in the photoproduction limit where we expect the largest contribution from genuine higher-order saturation contributions which could be accessed in ultraperipheral collisions of relativistic heavy ions.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Design of Graphene/Ionic Liquid Composites for Carbon Capture
Pore size is a crucial factor impacting gas separation in porous separation materials, but how to control the pore size to optimize the separation performance remains a challenge. Here, we propose a design of graphene/ionic liquid composites with tunable slit pore sizes, where cations and anions of ionic liquids are intercalated between graphene layers. By varying the sizes of the ions, we show from first-principles density functional theory calculations that the accessible pore size can be tuned from 3.4 to 6.0 Å. Grand canonical Monte Carlo simulations of gas sorption find that the composite materials possess high CO 2 uptake at room temperature and 1 bar (up to ~8.5 mmol/g). Further simulations of the sorption of gas mixtures reveal that high CO 2 /N 2 and CO 2 /CH 4 adsorption selectivities can be obtained when the accessible pore size is <5 Å. This work suggests a new strategy to achieve tunable pore sizes via the graphene/IL composites for highly selective CO 2 /N 2 and CO 2 /CH 4 adsorption.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗