321034 research outputs found
Sort by
Hypergraph p-Laplacian Equations for Data Interpolation and Semi-supervised Learning
No full textHypergraph learning with p-Laplacian regularization has attracted a lot of attention due to its flexibility in modeling higher-order relationships in data. This paper focuses on its fast numerical implementation, which is challenging due to the non-differentiability of the objective function and the non-uniqueness of the minimizer. We derive a hypergraph p-Laplacian equation from the subdifferential of the p-Laplacian regularization. A simplified equation that is mathematically well-posed and computationally efficient is proposed as an alternative. Numerical experiments verify that the simplified p-Laplacian equation suppresses spiky solutions in data interpolation and improves classification accuracy in semi-supervised learning. The remarkably low computational cost enables further applications
The influence of different core-ring intensity distributions on the capillary shape and resulting weld in laser welding of steel
No full textLaser welding was carried out using a co-axial core-ring beam, with independent control of the power to the coreand the ring. The welding process was observed using high-power X-rays and high-speed imaging equipment.Adjusting the relative powers of the core and the ring beams was found to have profound effects on the shape ofboth the capillary created and the weld melt pool. Moderate laser power densities in the ring were found toincrease the top diameter of the weld pool and reduce fluctuations in the welding process. High laser powerdensities in the ring led to the creation of a second capillary which largely acted independently of the capillarygenerated by the laser power in the core. At high power densities and welding speeds the two capillaries werealmost completely separated in the direction of travel
Periodic surface structure formation on silicon surfaces by irradiation of single and multiple femtosecond laser pulses
No full textLaser-induced periodic surface structures (LIPSS) were obtained on Si substrates using various number of ultrashort laser pulses with 266 fs pulse width at a 1030 nm wavelength. The obtained structures were investigated with SEM, EDS and optical spectroscopy techniques, then the obtained results were simulated by theoretical calculations. In the applied theoretical approach, the LIPSS formation is simulated by a semi-empirical model of interference of the incident laser beam with surface scattered electromagnetic waves generated at a rough surface, which includes excitation of surface plasmon polaritons (SPP) and accounts for the transient intrapulse changes in optical properties due to photo-excitation of a dense plasma of electron-hole pairs. The threshold for excitation of SPPs depends on the laser energy fluence, wavelength, polarization direction and the pulse number. We determine these thresholds from first principal calculation of the macroscopic dielectric function of the photo- excited solid
Intrinsic non-Gaussianity of ultra slow-roll inflation
No full textWe study the non-Gaussian tail of the curvature fluctuation, , in an inflationary scenario with a transient ultra slow-roll phase that generates a localized large enhancement of the spectrum of . To do so, we implement a numerical procedure that provides the probability distribution of order by order in perturbation theory. The non-Gaussianities of can be shown to arise from its non-linear relation to the inflaton fluctuations and from the intrinsic non-Gaussianities of the latter, which stem from its self interactions. We find that intrinsic non-Gaussianities, which have often been ignored to estimate the abundance of primordial black holes in this kind of scenario, are important. The relevance of the intrinsic contribution depends on the rapidity with which the transient ultra slow-roll phase occurs, as well as on its duration. Our method cannot be used accurately when the perturbative in-in formalism fails to apply, highlighting the relevance of developing fully non-perturbative approaches to the problem
Effect of Strain Rate on Deformation Mechanisms of a Mg Alloy
No full textThe effect of strain rate on dislocation slip and deformation twinning in a Mg–3wt.pctAl alloy was investigated by three-dimensional X-ray diffraction (3DXRD). In situ tensile tests were performed at strain rates of 10−4, 10−3, and 10−2 s−1. In each tested sample, more than 500 grains were indexed by 3DXRD, and their deformation was tracked. By measuring diffraction peak broadening, plastic deformation can be quantitatively analyzed in individual grains. Basal slip, prismatic slip, and pyramidal I slips were identified in various grains during deformation, and their critical resolved shear stress (CRSS) values were systematically evaluated. At higher strain rates, non-basal slip systems exhibited increased activity, although basal slip remained dominant. Twin nucleation was observed in a limited number of grains, including some with orientations unfavorable for twinning. Interestingly, in certain cases, the resolved shear stress on the activated twinning systems was negative, indicating a deviation from the generalized Schmid law. This phenomenon is likely associated with prior dislocation activity in the matrix grains, which could create localized conditions conducive to twin formation
Low-temperature luminescence of LiBO: Ag glass excited by synchrotron radiation
No full textThe paper presents the results of studies of photoluminescence (PL) of glass at a temperature of 12.6 K under the influence of synchrotron radiation λexc = 45 nm. It was found that the obtained PL spectra consist of three bands = 278, 336, and 379 nm. Studies of the excitation spectra by scanning with the registration of luminescent radiation at λmon = 280, 330, and 390 nm revealed four bands for each . It is proposed that the main band at 278 nm in the excitation spectra is associated with centers based on Ag ions that occupy the Li sites in the glass structure, while the other two bands at 336 and 379 nm are associated with centers based on Ag and Ag conglomerates. B – O centers based on structural defects in boron oxide complexes may be responsible for the weak 110 nm excitation band
Molecule effect in inner shell sequential doubleionization
No full textWe present a theoretical framework for the angular distribution of the two emitted photoelectrons in sequential inner shell two-photon double ionization (TPDI) of uniformly distributed molecules. As an example, we numerically calculated the sequential double ionization of the 4d shells of the iodine atom in CHI with a length and velocity gauge and obtained the corresponding angular anisotropy parameters. For comparison, we also compute the anisotropy parameters for atomic xenon. The anisotropy parameter for the second ionization of CHI is strongly influenced by the molecular structure, providing a sensitive probe for photoionization dynamics in polyatomic systems. This study provides a computational code for simulating sequential TPDI processes, enabling one to computationally study the molecular effects in the angular distribution for sequential inner-shell double ionization
Benchmarking CHGNet Universal Machine Learning Interatomic Potential against DFT and EXAFS: The Case of Layered WS and MoS
No full textUniversal machine learning interatomic potentials (uMLIPs) deliver near \emph{ab initio} accuracy in energy and force calculations at a low computational cost, making them invaluable for materials modeling. Although uMLIPs are pretrained on vast \emph{ab initio} data sets, rigorous validation remains essential for their ongoing adoption. In this study, we use the CHGNet uMLIP to model thermal disorder in isostructural layered 2H-WS and 2H-MoS, benchmarking it against \emph{ab initio} data and extended X-ray absorption fine structure (EXAFS) spectra, which capture thermal variations in bond lengths and angles. Fine-tuning CHGNet with compound-specific \emph{ab initio} (density functional theory (DFT)) data mitigates the systematic softening (i.e., force underestimation) typical of uMLIPs and simultaneously improves the alignment between molecular dynamics-derived and experimental EXAFS spectra. While fine-tuning with a single DFT structure is viable, using 100 structures is recommended to accurately reproduce EXAFS spectra and achieve DFT-level accuracy. Benchmarking the CHGNet uMLIP against both DFT and experimental EXAFS data reinforces confidence in its performance and provides guidance for determining optimal fine-tuning data set sizes
Importance of local tetraquark operators for
No full textThe doubly charmed tetraquark observed at LHCb has attracted considerable interest in recent years. To accurately determine its finite-volume spectrum, a variational analysis using a large basis of operators, including bilocal scattering operators, but also local tetraquark operators, should be employed. Using Wilson-clover fermions at the -flavour-symmetric point, we investigated the importance of local tetraquark operators for the spectrum by adding them to a large basis of bilocal and scattering operators. We performed this calculation using the distillation framework combined with a position-space sampling method that we recently developed. This method makes local tetraquark operators affordable in distillation. Upon including local tetraquark operators, we observe significant shifts in the estimates of several energy levels. Finally, we show the effect of these shifts on the scattering phase shifts obtained from a single-channel -wave Lüscher analysis