On the application of the residual method for the correction of inconsistent problems of convex programming

For the correction of a convex programming problem with potentially inconsistent constraint system (an improper problem), we apply the residual method, which is a standard regularization procedure for ill-posed optimization models. A problem statement typical for the residual method is reduced to a minimization problem for an appropriate penalty function. We apply two classical penalty functions: the quadratic penalty function and the exact Eremin-Zangwill penalty function. For each of the approaches, we establish convergence conditions and bounds for the approximation error.     

Application of the penalty method to nonstationary approximation of an optimization problem

We solve a general optimization problem, where only approximation sequences are known instead of exact values of the goal function and feasible set. Under these conditions we suggest to utilize a penalty function method. We show that its convergence is attained for rather arbitrary means of approximation via suitable coercivity type conditions.     

A power penalty approach to a mixed quasilinear elliptic complementarity problem

In this paper, a power penalty approximation method is proposed for solving a mixed quasilinear elliptic complementarity problem. The mixed complementarity problem is first reformulated as a double obstacle quasilinear elliptic variational inequality problem. A nonlinear elliptic partial differential equation is then defined to approximate the resulting variational inequality by using a power penalty approach. The existence and uniqueness of the solution to the partial differential penalty equation are proved. It is shown that, under some mild assumptions, the sequence of solutions to the penalty equations converges to the unique solution of the variational inequality problem as the penalty parameter tends to infinity. The error estimates of the convergence of this penalty approach are also derived. At last, numerical experimental results are presented to show that the power penalty approximation method is efficient and robust.

     

"准"精确惩罚函数法的渐近性分析

1引言众所周知,罚函数法在最优化理论与数值计算中占据着极其重要的位置,作为求解约束优化问题的一类重要方法,在上世纪五、六十年代曾经历一次发展高潮．近十几年来,伴随着对数障碍函数法在内点法中取得的成功,罚函数法的研究又呈现出一个小高潮[2,3,4]．在罚函数方法里,精确惩罚函数法有着非常吸引人的性质,即,当罚参数大于某个有限门槛值时,仅通过求解单个无约束罚问题便可得到原问题的最优解,从而省去了一般罚函数法解系列无约束优化问题的工作量．     

An adaptive extrapolation discontinuous Galerkin method for the valuation of Asian options

We consider the approximation of the optimal stopping problem associated with ultradiffusion processes in the context of mathematical finance and the valuation of Asian options. In particular, the value function is characterized as the solution of an ultraparabolic variational inequality. Employing the penalty method and a regularization of the state space, we develop higher-order adaptive approximation schemes which utilize the extrapolation discontinuous Galerkin method in temporal space. Numerical examples are provided in order to demonstrate the approach.     

Two-level iterative method for non-stationary mixed variational inequalities

We consider a mixed variational inequality problem involving a set-valued nonmonotone mapping and a general convex function, where only approximation sequences are known instead of exact values of the cost mapping and function, and feasible set. We suggest to apply a two-level approach with inexact solutions of each particular problem with a descent method and partial penalization and evaluation of accuracy with the help of a gap function. Its convergence is attained without concordance of penalty, accuracy, and approximation parameters under coercivity type conditions.     

CVaR-constrained stochastic programming reformulation for stochastic nonlinear complementarity problems

We reformulate a stochastic nonlinear complementarity problem as a stochastic programming problem which minimizes an expected residual defined by a restricted NCP function with nonnegative constraints and CVaR constraints which guarantee the stochastic nonlinear function being nonnegative with a high probability. By applying smoothing technique and penalty method, we propose a penalized smoothing sample average approximation algorithm to solve the CVaR-constrained stochastic programming. We show that the optimal solution of the penalized smoothing sample average approximation problem converges to the solution of the corresponding nonsmooth CVaR-constrained stochastic programming problem almost surely. Finally, we report some preliminary numerical test results.     

���ǽ��׷��õĶ������ռ��Ͷ����Ϸ��տ���ģ��

In this paper, we introduced a transaction costs function and established a portfolio model of risk management with second stochastic dominance constraints. This model does not need to make any assumptions about the utility function of the investors and the distribution of the risk assets income, and it can ensure that the choices of the risk-averse investor can be randomly better than a reference value, so it can avoid the high risk investment. We provide a smoothing penalty sample average approximation method for solving this optimization problem. We prove that the smoothing penalty problem is equivalent to the original problem. Numerical results prove that the model and the method are efficient.     

Two-level iteration penalty and variational multiscale method for steady incompressible flows

In this paper, we study two-level iteration penalty and variational multiscale method for the approximation of steady Navier-Stokes equations at high Reynolds number. Comparing with classical penalty method, this new method does not require very small penalty parameter $\varepsilon$. Moreover, two-level mesh method can save a large amount of CPU time. The error estimates in $H^1$ norm for velocity and in $L^2$ norm for pressure are derived. Finally, two numerical experiments are shown to support the efficiency of this new method.     

A class of generalized variable penalty methods for nonlinear programming

A class of generalized variable penalty formulations for solving nonlinear programming problems is presented. The method poses a sequence of unconstrained optimization problems with mechanisms to control the quality of the approximation for the Hessian matrix, which is expressed in terms of the constraint functions and their first derivatives. The unconstrained problems are solved using a modified Newton's algorithm. The method is particularly applicable to solution techniques where an approximate analysis step has to be used (e.g., constraint approximations, etc.), which often results in the violation of the constraints. The generalized penalty formulation contains two floating parameters, which are used to meet the penalty requirements and to control the errors in the approximation of the Hessian matrix. A third parameter is used to vary the class of standard barrier or quasibarrier functions, forming a branch of the variable penalty formulation. Several possibilities for choosing such floating parameters are discussed. The numerical effectiveness of this algorithm is demonstrated on a relatively large set of test examples.The author is thankful for the constructive suggestions of the referees.     

On the convergence of solutions for a class of nonconformal FEM schemes for quasilinear elliptic equations

We consider versions of the nonconformal finite element method for the approximation to a second-order quasilinear elliptic equation in divergence form. For the construction of grid schemes, we use an approach used earlier for the nonstationary convection-diffusion equation and based on the Galerkin-Petrov limit approximation to the mixed statement of the original problem. The accuracy of solutions of nonconformal schemes with triangular linear finite elements is estimated in the absence of interior penalty terms, which are usually used in methods close to DG-methods for the stabilization of the scheme solution.     

A Quasi-Newton Quadratic Penalty Method for Minimization Subject to Nonlinear Equality Constraints

We present a modified quadratic penalty function method for equality constrained optimization problems. The pivotal feature of our algorithm is that at every iterate we invoke a special change of variables to improve the ability of the algorithm to follow the constraint level sets. This change of variables gives rise to a suitable block diagonal approximation to the Hessian which is then used to construct a quasi-Newton method. We show that the complete algorithm is globally convergent. Preliminary computational results are reported.     

曲率障碍下四阶变分不等式的交替方向乘子法北大核心CSCD

对于重调和算子和曲率障碍表示的变分不等式,提出了自适应交替方向乘子数值解法(SADMM).对问题引入一个辅助变量表示曲率函数的增广Lagrange函数,导出一个约束极小值问题,并且该问题等价于一个鞍点问题.然后采用交替方向乘子法(ADMM)求解这个鞍点问题.通过采用平衡原理和迭代函数,得到了自动调整罚参数的自适应法则,从而提高了计算效率.证明了该方法的收敛性,并给出了利用迭代函数近似罚参数的具体方法.最后,用数值计算结果验证了该方法的有效性.     

Solving stochastic mathematical programs with equilibrium constraints via approximation and smoothing implicit programming with penalization

In this paper, we consider the stochastic mathematical programs with linear complementarity constraints, which include two kinds of models called here-and-now and lower-level wait-and-see problems. We present a combined smoothing implicit programming and penalty method for the problems with a finite sample space. Then, we suggest a quasi-Monte Carlo approximation method for solving a problem with continuous random variables. A comprehensive convergence theory is included as well. We further report numerical results with the so-called picnic vender decision problem.     

低阶精确罚函数的一种光滑化逼近

本文对不等式约束优化问题给出了低阶精确罚函数的一种光滑化逼近.提出了通过搜索光滑化后的罚问题的全局解而得到原优化问题的近似全局解的算法.给出了几个数值例子以说明所提出的光滑化方法的有效性.     

Vector scheduling with rejection on a single machine

In this paper, we study a vector scheduling problem with rejection on a single machine, in which each job is characterized by a d-dimension vector and a penalty, in the sense that, jobs can be either rejected by paying a certain penalty or assigned to the machine. The objective is to minimize the sum of the maximum load over all dimensions of the total vector of all accepted jobs, and the total penalty of rejected jobs. We prove that the problem is NP-hard and design two approximation algorithms running in polynomial time. When d is a fixed constant, we present a fully polynomial time approximation scheme.     

Hilbert空间的一个约束极值问题和非协调有限元的加罚方法

在实践中成功地运用了非协调有限元并因此而使它受到人们的注意和研究。使用非协调元的一个方法是加罚方法。这是Babuska和Zlamal首先在[1]中提出的。冯康在[4]中证明:一般的,当惩罚项满足某些条件时,加罚方法总是收敛的。 非协调元(以及杂交元和某些其他数值解法)的理论研究,可以归结为Hilbert空间的一个约束极值问题。本文首先对这一抽象问题进行了分析,证明了加罚方法的收敛性     

关于年龄相关的种群扩散系统的最优边界控制计算的惩罚移位法

研究了惩罚移位法应用于一类种群扩散系统最优边界控制的计算,构造了其逼近程序,并证明了这种方法的收敛性.     

带次模惩罚的优先设施选址问题的近似算法

研究带次模惩罚的优先设施选址问题, 每个顾客都有一定的服务水平要求, 开设的设施只有满足了顾客的服务水平要求, 才能为顾客提供服务, 没被服务的顾客对应一定的次模惩罚费用. 目标是使得开设费用、连接费用与次模惩罚费用之和最小. 给出该问题的整数规划、 线性规划松弛及其对偶规划. 基于原始对偶和贪婪增广技巧, 给出该问题的两个近似算法, 得到的近似比分别为3和2.375.     

对流占优的Sobolev方程的投影稳定化有限元方法

对于对流占优的Sobolev方程,提出了一种新的投影稳定化有限元方法,建立了半离散和全离散的投影稳定化格式,给出了解的稳定性和收敛性分析.该方法能够有效克服对流占优,与内罚方法相比,投影格式更简单,计算量更小,且得到的C—N格式是无条件稳定的,时间精度达到了二阶.最后,通过实验证明,数值结果与理论结果完全一致.