希尔伯特空间有限区域成像积分算子的谱

本文讨论了希尔伯特空间中有限区域成像积分算子的本征值和本征函数,研究了非对称核的情形,并用微扰和有限秩方法计算了衍射受限和离焦情况下的本征值和本征函数。     

On the Accuracy of Gaussian Approximation in Hilbert Space

The bound on remainder term in CLT for a sum of independent random variables taking values in Hilbert space is obtained. This bound sharpens the result due to Bentkus and Götze.     

The Hilbert problem: The case of infinitely many discontinuity points of coefficients

We obtain a solution to the Hilbert boundary value problem in the theory of analytic functions on the half-plane in the case that the coefficients of the boundary condition have countably many discontinuity points of the first kind. We elaborate the two substantially different situations: the series consisting of the jumps of the argument of the coefficient function and the increments of its continuous part converges and this series diverges. Accordingly, Hilbert problems with finite and infinite indices result. We derive formulas for the general solution and investigate the pictures of solvability of these problems.     

Epidifferentiability of the map of infima in Hilbert spaces

This article deals with derivatives for set-valued maps that take values in ordered vector spaces, in particular it concerns about the relationship between the epiderivatives of a set-valued map and its associated map of infima. When the image space is a real separable Hilbert space ordered by an orthonormal basis, by using a variational technique based on a decoupling of the ordering cone into half-spaces, we show that both epiderivatives coincide under certain hypothesis of compactness and stability. Furthermore we obtain some computation formulas for these derivatives in terms of associated scalar set-valued maps.     

Approximate solution of the Fredholm integral equation of the first kind in a reproducing kernel Hilbert space

An approach for solving Fredholm integral equations of the first kind is proposed for in a reproducing kernel Hilbert space (RKHS). The interest in this problem is strongly motivated by applications to actual prospecting. In many applications one is puzzled by an ill-posed problem in space C[a,b] or L²[a,b], namely, measurements of the experimental data can result in unbounded errors of solutions of the equation. In this work, the representation of solutions for Fredholm integral equations of the first kind is obtained if there are solutions and the stability of solutions is discussed in RKHS. At the same time, a conclusion is obtained that approximate solutions are also stable with respect to _∞ or _L² in RKHS. A numerical experiment shows that the method given in the work is valid.     

Modeling of non-stationary ground motion using the mean reverting stochastic process

This paper presents a new method for modeling amplitude and frequency non-stationary earthquake ground motions using a scalar first order dynamic mean reverting stochastic differential equation driven by Brownian motion with parametric time varying coefficients. It determines the proper relationship between these time varying parametric coefficients and presents the statistical and probability distribution characteristics of the response solution. It demonstrates the applicability of the method by presenting some simulations of amplitude and frequency non-stationary earthquake ground motions. The verification of the amplitude and frequency non-stationary contents of the mean reverting stochastic ground motions is demonstrated using the Hilbert–Huang transform method. Also a corresponding interpretation between the coefficients of the proposed model and the coefficients of the usual oscillatory second order differential equation driven by white Gaussian noise is presented along with some comments how it can be applied to simulate ground motions consistent with acceleration target records such as boxcar, trapezoidal, other exponential functions, or compound and target records at source, near field, and far field distances.     

Rational approximation to trigonometric operators

We consider the approximation of trigonometric operator functions that arise in the numerical solution of wave equations by trigonometric integrators. It is well known that Krylov subspace methods for matrix functions without exponential decay show superlinear convergence behavior if the number of steps is larger than the norm of the operator. Thus, Krylov approximations may fail to converge for unbounded operators. In this paper, we propose and analyze a rational Krylov subspace method which converges not only for finite element or finite difference approximations to differential operators but even for abstract, unbounded operators. In contrast to standard Krylov methods, the convergence will be independent of the norm of the operator and thus of its spatial discretization. We will discuss efficient implementations for finite element discretizations and illustrate our analysis with numerical experiments. AMS subject classification (2000)  65F10, 65L60, 65M60, 65N22     

带有Radon-Nikodym导数的算子值自由Fisher信息量

本文研究了带有Radon-Nikodym导数的算子值自由Fisher信息量.利用模框架理论,得到了一个半圆元和一个子代数之间的合并自由关系,推广了D.Voiculescu等人的工作.     

On box totally dual integral polyhedra

Edmonds and Giles introduced the class of box totally dual integral polyhedra as a generalization of submodular flow polyhedra. In this paper a geometric characterization of these polyhedra is given. This geometric result is used to show that each TDI defining system for a box TDI polyhedron is in fact a box TDI system, that the class of box TDI polyhedra is in co-NP and is closed under taking projections and dominants, that the class of box perfect graphs is in co-NP, and a result of Edmonds and Giles which is related to the facets of box TDI polyhdera.Supported by a grant from the Alexander von Humboldt-Stiftung.     

On the greedy solution in integer linear programming

A version of the greedy method not using any knapsack relaxation of the integer programming problem is considered in this paper. It is based on a natural partial ordering of the vectors. Our aim is to determine a large class of problems where the greedy solution is always optimal. The results generalize some theorems of an early paper of Magazine, Nemhauser and Trotter and at the same time show a connection between two different notions of combinatorics: the greedy method and the Hilbert basis.

---

Zusammenfassung In dieser Arbeit wird eine Version des Greedy-Algorithmus zur Lösung ganzzahliger linearer Optimierungsprobleme benutzt, die kein Rucksackproblem als Relaxation verwendet. Das Verfahren basiert auf der natürlichen partiellen Ordnung von Vektoren. Ziel der Arbeit ist es, eine möglichst große Problemklasse zu beschreiben, für die die Greedy-Lösung optimal ist. Die Ergebnisse verallgemeinern Sätze einer früheren Arbeit von Magazine, Nemhauser und Trotter und zeigen gleichzeitig einen Bezug zwischen zwei verschiedenen Gebieten der Kombinatorik auf: des Greedy-Verfahrens und von Hubert-Basen.