The generalized quasilinearization method for second-order boundary value problem has been extended when the forcing function is the sum of two functions without require that any of the two functions involved to be 2-hyperconvex or 2-hyperconcave. Two sequences are developed under suitable conditions which converge to the unique solution of the boundary value problem. Furthermore, the convergen...

We give bounds on the rate of convergence to equilibrium of the symmetric simple exclusion process in Z d. Our results include the existent results in the literature. We get better bounds and larger class of initial states via a unified approach. The method includes a comparison of the evolution of n interacting particles with n independent ones along the whole time trajectory.

The leading term in the normal approximation to the distribution of Student’s t statistic is derived in a general setting, with the sole assumption being that the sampled distribution is in the domain of attraction of a normal law. The form of the leading term is shown to have its origin in the way in which extreme data influence properties of the Studentized sum. The leadingterm approximation ...

In this article, we prove stability of strong detonation waves and find their rate of convergence for a combustion model. Our results read as follows: I) There exists a global solution that converges exponentially in time to a strong detonation wave, provided that the initial data is a small perturbation of a strong detonation wave that decays exponentially in |x|. II) When the initial perturba...

The existing stability analysis of particle swarm optimization (PSO) algorithm is chiefly concluded by the assumption of constant transfer matrix or time-varying random transfer matrix. Firstly, one counterexample is provided to show that the existing convergence analysis is not possibly valid for PSO system involving random variables. Secondly, the joint spectral radius, mainly calculated by t...

With explosion of data size and limited storage space at a single location, data are often distributed at different locations. We thus face the challenge of performing largescale machine learning from these distributed data through communication networks. In this paper, we study how the network communication constraints will impact the convergence speed of distributed machine learning optimizat...

length-biased data are widely seen in applications. they are mostly applicable in epidemiological studies or survival analysis in medical researches. here we aim to propose a berry-esseen type bound for the kernel density estimator of this kind of data.the rate of normal convergence in the proposed berry-esseen type theorem is shown to be o(n^(-1/6) ) modulo logarithmic term as n tends to infin...

In this paper, we show a phenomenon, which we named “super-convergence”, where residual networks can be trained using an order of magnitude fewer iterations than is used with standard training methods. One of the key elements of superconvergence is training with cyclical learning rates and a large maximum learning rate. Furthermore, we present evidence that training with large learning rates im...

Robust estimators of location and dispersion are often used in the elliptical model to obtain an uncontaminated and highly representative subsample by trimming the data outside an ellipsoid based in the associated Mahalanobis distance. Here we analyze some one (or k)-step Maximum Likelihood Estimators computed on a subsample obtained with such a procedure. We introduce different models which ar...