Problems and Solutions in Heterogeneous Cooperative Learning

Heterogeneous cooperative learning,as the best kind of cooperative learning,is quite often adopted to make utmost use of the human power in second language learning.On the other hand,some problems existing in it make it far from satisfactory.In this thesis,the writer will analyse the problems and put forward some solutions to perfect HCL.     

Accuracy of suboptimal solutions to kernel principal component analysis

For Principal Component Analysis in Reproducing Kernel Hilbert Spaces (KPCA), optimization over sets containing only linear combinations of all n-tuples of kernel functions is investigated, where n is a positive integer smaller than the number of data. Upper bounds on the accuracy in approximating the optimal solution, achievable without restrictions on the number of kernel functions, are derived. The rates of decrease of the upper bounds for increasing number n of kernel functions are given by the summation of two terms, one proportional to n ^−1/2 and the other to n ⁻¹, and depend on the maximum eigenvalue of the Gram matrix of the kernel with respect to the data. Primal and dual formulations of KPCA are considered. The estimates provide insights into the effectiveness of sparse KPCA techniques, aimed at reducing the computational costs of expansions in terms of kernel units.     

Monomiality and partial differential equations

We show that the combination of the formalism underlying the principle of monomiality and of methods of an algebraic nature allows the solution of different families of partial differential equations. Here we use different realizations of the Heisenberg–Weyl algebra and show that a Sheffer type realization leads to the extension of the method to finite difference and integro-differential equations.     

湖南省居住物业管理现状与问题剖析

分析了湖南省居住物业管理现状,对目前湖南省居住物业管理存在的问题进行了比较全面的剖析.     

An algorithmic framework for convex mixed integer nonlinear programs

This paper is motivated by the fact that mixed integer nonlinear programming is an important and difficult area for which there is a need for developing new methods and software for solving large-scale problems. Moreover, both fundamental building blocks, namely mixed integer linear programming and nonlinear programming, have seen considerable and steady progress in recent years. Wishing to exploit expertise in these areas as well as on previous work in mixed integer nonlinear programming, this work represents the first step in an ongoing and ambitious project within an open-source environment. COIN-OR is our chosen environment for the development of the optimization software. A class of hybrid algorithms, of which branch-and-bound and polyhedral outer approximation are the two extreme cases, are proposed and implemented. Computational results that demonstrate the effectiveness of this framework are reported. Both the library of mixed integer nonlinear problems that exhibit convex continuous relaxations, on which the experiments are carried out, and a version of the software used are publicly available.     

不适定问题的基本定理（Ⅱ）

前[1]已证基本定理关于条件(B)与条件(I—D)的等价性，本给出定理中条件(I)与条件(I-D)的等价性，从而三个条件(B)、(I)、(I—D)是相互等价的。     

Numerical techniques for finding equilibrium in von Neumann’s model

Numerical methods for finding an equilibrium position in von Neumann’s multisector economy model are proposed. They can be implemented using computers with floating-point arithmetic. The proposed methods are based on reducing the problem to solving the corresponding matrix games.     

Using discrete optimization algorithms to find minimum energy configurations of slender cantilever beams with non-convex energy functions

This paper deals with a new solution technique for approximately solving certain variational problems in elasticity by using discrete optimization techniques that were originally used in information theory. This allows us to easily and approximately solve large deformation buckling problems for slender cantilever beams (including post-buckling behavior) as well as problems where the strain energy function is non-convex.The core idea is to quantize or discretize the variables describing the possible configurations of the body. This, when combined with the fact that the variational problem has an inherent Markov structure allows us to use computationally efficient search techniques based on dynamic programming (equivalent to finding the shortest path in a weighted directed graph) to find optimal solutions within the quantized state space. The results can be used in two ways: (1) directly as a fast approximate solution to the variational problem (2) As a means for finding very good (nearly minimum energy) initial configurations for application of conventional minimization techniques, which might otherwise fail because of a poor starting configurations which are far from the global minimum. We demonstrate both these uses in the paper.     

On non-Newtonian p-fluids. The pseudo-plastic case

In the following we study a class of stationary Navier-Stokes equations with shear dependent viscosity, under the non-slip (Dirichlet) boundary condition. We consider pseudo-plastic fluids. A fluid is said pseudo-plastic, or shear thinning, if in Eq. (1.1) below one has p<2. We are interested in global (i.e., up to the boundary) regularity results, in dimension n=3, for the second order derivatives of the velocity and the first order derivatives of the pressure. We consider a cubic domain Ω and impose the non-slip boundary condition only on two opposite faces. On the other faces we assume periodicity, as a device to avoid effective boundary conditions. This choice is made so that we work in a bounded domain Ω and simultaneously with a flat boundary.     

Optimal control of heterogeneous systems: Basic theory

A general model of a heterogeneous control system is introduced in the form of a first order distributed system with nonlocal dynamics and exogenous side-conditions. The heterogeneity is represented by a parameter taking values in an abstract measurable space, so that both continuous and discrete heterogeneity, as well as probabilistic heterogeneity without density, are included. A distributed and a lumped controls are involved, the latter appearing also in the side conditions. An existence theorem is proved for the uncontrolled system, and the sensitivity of the solution with respect to the control variables is estimated. The main result is an optimality condition in the form of the Pontryagin local maximum principle. A global maximum principle holds for the distributed control under an additional condition that rules out discrete measurable heterogeneity spaces. A number of possible applications are outlined: age-structured systems, size-structured systems, (nonlocal) advection-reaction equations, static parametric heterogeneity in epidemiology, and two-stage control systems with uncertain switching time.