Journal of Computational Mathematics and Data Science最新文献

The recent investigations ensure that, the effect of an endoscope on the peristaltic flow is very important for medical diagnosis and it has many clinical applications such as gastric juice motion in the small intestine when an endoscope is inserted through it. In the current article, the influence of magnetohydrodynamic (MHD) on the peristaltic propulsion of non-Newtonian fluid (considered as couple stress fluid) in a tube consisting of endoscope has been considered. The couple stress fluid occupies the space between two co-axial inclined tubes. The inner tube is uniformly circular and rigid while the outer tube considered as sinusoidal wave. The fluid motion is discussed in a wave frame which is moving with the constant velocity. The governing equations of two-dimensional flow have been abridged under the lubrication approach. Analytical solutions have been obtained for the velocity and pressure gradient with the help of modified Bessel functions. Numerical integration is used to evaluate the pressure difference and friction forces. The effect of emerging flow parameters on the velocity, frictional forces, pressure difference, pressure gradient and trapping phenomenon have been discussed. It is noted that, the magnetic force resists the flow and pumping rate in the peristaltic flow enhances from the horizontal to vertical tube. The present study has a wide range of applications in bio-medical engineering like the transport phenomenon in peristaltic micro pumps.

Strongly interacting electrons in solids are generically described by Hubbard-type models, and the impact of solar light can be modeled by an additional time-dependence. This yields a finite dimensional system of ordinary differential equations (ODE)s of Schrödinger type, which can be solved numerically by exponential time integrators of Magnus type. The efficiency may be enhanced by combining these with operator splittings. We will discuss several different approaches of employing exponential-based methods in conjunction with an adaptive Lanczos method for the evaluation of matrix exponentials and compare their accuracy and efficiency. For each integrator, we use defect-based local error estimators to enable adaptive time-stepping. This serves to reliably control the approximation error and reduce the computational effort.

Recently some neural networks have been proposed for computing approximate solutions to partial differential equations. For second order elliptic or parabolic PDEs, this is possible by using penalized Least squares formulations of PDEs. In this article, for some second order elliptic PDEs we propose a theoretical setting, and we investigate the abstract convergence results between the solution and the computed one with neural networks. These results are obtained by minimizing appropriate loss functions made of a least squares formulation of the PDE augmented with a penalization term for accounting the Dirichlet boundary conditions. More precisely, it is shown that the error has two components, one due to the neural network and one due to the way the boundary conditions are imposed (via a penalization technic). The interplay between the two errors shows that the accuracy of the neural network has to be chosen accordingly with the accuracy of the boundary conditions.

In this paper, we present a new family of distributions based on a particular case of a transform introduced by Yun (2014). Among others, this transform demonstrates great flexibility and nice mathematical properties which can be useful in a statistical context (continuous derivatives of all order, simplicity of the inverse transform, etc.). We propose a new three-parameter distribution from this family, namely the Yun–Weibull (YW) distribution. Some statistical properties of this distribution are studied, involving flexible hazard rate shapes. Subsequently, the statistical inference of the YW distribution is investigated. The parameters are estimated by employing the maximum likelihood estimation method. We establish the existence and uniqueness of the obtained estimators. The YW distribution is applied to fit two practical data sets. As a main result of our analysis, the new distribution is found to be more appropriate to these data sets than other competitive distributions. Moreover, the uniqueness of the parameter estimates of the YW distribution is studied using the profile log-likelihood function visually under the two practical data sets.

Gabor functions have wide-spread applications both in analyzing the visual cortex of mammalians and in designing machine vision algorithms. It is known that the receptive field of neurons of V1 layer in the visual cortex can be accurately modeled by Gabor functions. In addition, Gabor functions are extensively used for feature extraction in machine vision tasks. In this paper, we prove that Gabor functions are translation-invariant positive-definite kernels and show that the problem of image representation with Gabor functions can be formulated as infinite kernel learning regression. Specifically, we use the stabilized infinite kernel learning regression algorithm that has already been introduced for learning translation-invariant positive-definite kernels and has enough flexibility and generality to embrace the class of Gabor kernels. The algorithm yields a representation of the image as a support vector expansion with a compound kernel that is a finite mixture of Gabor functions. The problem with this representation is that all Gabor functions are present at all support vector pixels. Using LASSO, we propose a method for sparse representation of an image with Gabor functions in which each Gabor function is positioned at a very sparse set of pixels. As a practical application, we introduce a novel method for learning a dataset-specific set of Gabor filters that can be used subsequently for feature extraction. Our experiments on CMU-PIE and Extended Yale B datasets show that use of the learned Gabor filters significantly improves the recognition accuracy of a recently introduced face recognition algorithm.

The performance of exponential-based numerical integrators for the time propagation of the equations associated with the multiconfiguration time-dependent Hartree–Fock (MCTDHF) method for the approximation of the multi-particle Schrödinger equation in one space dimension is assessed. Among the most popular integrators such as Runge–Kutta methods, time-splitting, exponential integrators and Lawson methods, exponential Lawson multistep methods with one predictor–corrector step provide the best stability and accuracy at the least effort. This assessment is based on the observation that the evaluation of the nonlocal terms associated with the potential is the computationally most demanding part of such a calculation in our setting. In addition, the predictor step provides an estimator for the local time-stepping error, thus allowing for adaptive time-stepping which reflects the smoothness of the solution and enables to reliably control the accuracy of a computation in a robust way, without the need to guess an optimal stepsize a priori. One-dimensional model examples are studied to compare different time integrators and demonstrate the successful application of our adaptive methods.

In this paper we discuss the potential of using artificial neural networks as smooth priors in classical methods for inverse problems for PDEs. Exploring that neural networks are global and smooth function approximators, the idea is that neural networks could act as attractive priors for the coefficients to be estimated from noisy data. We illustrate the capabilities of neural networks in the context of the Poisson equation and we show that the neural network approach show robustness with respect to noisy, incomplete data and with respect to mesh and geometry.