An inexact interior point method for L 1-regularized sparse covariance selection

Sparse covariance selection problems can be formulated as log-determinant (log-det) semidefinite programming (SDP) problems with large numbers of linear constraints. Standard primal–dual interior-point methods that are based on solving the Schur complement equation would encounter severe computational bottlenecks if they are applied to solve these SDPs. In this paper, we consider a customized inexact primal–dual path-following interior-point algorithm for solving large scale log-det SDP problems arising from sparse covariance selection problems. Our inexact algorithm solves the large and ill-conditioned linear system of equations in each iteration by a preconditioned iterative solver. By exploiting the structures in sparse covariance selection problems, we are able to design highly effective preconditioners to efficiently solve the large and ill-conditioned linear systems. Numerical experiments on both synthetic and real covariance selection problems show that our algorithm is highly efficient and outperforms other existing algorithms.     

Inexact non-interior continuation method for solving large-scale monotone SDCP

For exact Newton method for solving monotone semidefinite complementarity problems (SDCP), one needs to exactly solve a linear system of equations at each iteration. For problems of large size, solving the linear system of equations exactly can be very expensive. In this paper, we propose a new inexact smoothing/continuation algorithm for solution of large-scale monotone SDCP. At each iteration the corresponding linear system of equations is solved only approximately. Under mild assumptions, the algorithm is shown to be both globally and superlinearly convergent.     

A new sequential systems of linear equations algorithm of feasible descent for inequality constrained optimization

Based on a new efficient identification technique of active constraints introduced in this paper, a new sequential systems of linear equations （SSLE） algorithm generating feasible iterates is proposed for solving nonlinear optimization problems with inequality constraints. In this paper, we introduce a new technique for constructing the system of linear equations, which recurs to a perturbation for the gradients of the constraint functions. At each iteration of the new algorithm, a feasible descent direction is obtained by solving only one system of linear equations without doing convex combination. To ensure the global convergence and avoid the Maratos effect, the algorithm needs to solve two additional reduced systems of linear equations with the same coefficient matrix after finite iterations. The proposed algorithm is proved to be globally and superlinearly convergent under some mild conditions. What distinguishes this algorithm from the previous feasible SSLE algorithms is that an improving direction is obtained easily and the computation cost of generating a new iterate is reduced. Finally, a preliminary implementation has been tested.     

Non-Interior Continuation Method for Solving the Monotone Semidefinite Complementarity Problem

Recently, Chen and Tseng extended non-interior continuation/ smooth- ing methods for solving linear/ nonlinear complementarity problems to semidefinite complementarity problems (SDCP). In this paper we propose a non-interior continuation method for solving the monotone SDCP based on the smoothed Fischer—Burmeister function, which is shown to be globally linearly and locally quadratically convergent under suitable assumptions. Our algorithm needs at most to solve a linear system of equations at each iteration. In addition, in our analysis on global linear convergence of the algorithm, we need not use the assumption that the Fréchet derivative of the function involved in the SDCP is Lipschitz continuous. For non-interior continuation/ smoothing methods for solving the nonlinear complementarity problem, such an assumption has been used widely in the literature in order to achieve global linear convergence results of the algorithms.     

Non-Interior Continuation Method for Solving the Monotone Semidefinite Complementarity Problem

Recently, Chen and Tseng extended non-interior continuation/ smooth- ing methods for solving linear/ nonlinear complementarity problems to semidefinite complementarity problems (SDCP). In this paper we propose a non-interior continuation method for solving the monotone SDCP based on the smoothed Fischer—Burmeister function, which is shown to be globally linearly and locally quadratically convergent under suitable assumptions. Our algorithm needs at most to solve a linear system of equations at each iteration. In addition, in our analysis on global linear convergence of the algorithm, we need not use the assumption that the Fréchet derivative of the function involved in the SDCP is Lipschitz continuous. For non-interior continuation/ smoothing methods for solving the nonlinear complementarity problem, such an assumption has been used widely in the literature in order to achieve global linear convergence results of the algorithms.     

一类特征值反问题(IEP)的基于矩阵方程的Ulm型算法

应用求解算子方程的Ulm方法构造了求解一类矩阵特征值反问题（IEP）的新算法.所给算法避免了文献[Aishima K.，A quadratically convergent algorithm based on matrix equations for inverse eigenvalue problems，Linear Algebra and its Applications，2018，542：310-33]中算法在每次迭代中要求解一个线性方程组的不足，证明了在给定谱数据互不相同的条件下所给算法具有根收敛意义下的二次收敛性.数值实验表明本文所给算法在矩阵阶数较大时计算效果优于上文所给算法.     

Inexact non-interior continuation method for monotone semidefinite complementarity problems

Chen and Tseng (Math Program 95:431?C474, 2003) extended non-interior continuation methods for solving linear and nonlinear complementarity problems to semidefinite complementarity problems (SDCP), in which a system of linear equations is exactly solved at each iteration. However, for problems of large size, solving the linear system of equations exactly can be very expensive. In this paper, we propose a version of one of the non-interior continuation methods for monotone SDCP presented by Chen and Tseng that incorporates inexactness into the linear system solves. Only one system of linear equations is inexactly solved at each iteration. The global convergence and local superlinear convergence properties of the method are given under mild conditions.     

A TRULY GLOBALLY CONVERGENT FEASIBLE NEWTON-TYPE METHOD FOR MIXED COMPLEMENTARITY PROBLEMS

Typical solution methods for solving mixed complementarity problems either generatefeasible iterates but have to solve relatively complicated subproblems such as quadraticprograms or linear complementarity problems,or(those methods)have relatively simplesubproblems such as system of linear equations but possibly generate infeasible iterates.In this paper,we propose a new Newton-type method for solving monotone mixed com-plementarity problems,which ensures to generate feasible iterates,and only has to solve asystem of well-conditioned linear equations with reduced dimension per iteration.Withoutany regularity assumption,we prove that the whole sequence of iterates converges to a so-lution of the problem(truly globally convergent).Furthermore,under suitable conditions,the local superlinear rate of convergence is also established.     

Superlinear/quadratic smoothing Broyden-like method for the generalized nonlinear complementarity problem

In this article, we first reformulate the generalized nonlinear complementarity problem (GNCP) over a polyhedral cone as a smoothing system of equations and then suggest a smoothing Broyden-like method for solving it. The proposed algorithm has to solve only one system of nonhomogeneous linear equations, perform only one line search and update only one matrix per iteration. We show that the iteration sequence generated by the proposed algorithm converges globally and superlinearly under suitable conditions. Furthermore, the algorithm has local quadratic convergence under mild assumptions. Some numerical examples are given to illustrate the performance and efficiency of the presented algorithm.     

P_0函数非线性互补问题的一步非内点连续方法的收敛性

本文对于Ｐ０函数非线性互补问题提出了一个基于Ｋａｎｚｏｗ光滑函数的一步非内点连续方法,在适当的假设条件下,证明了方法的全局线性及局部二次收敛性．特别,在方法的全局线性收敛性的分析中,不需要假定非线性互补问题的函数的Ｊａｃｏｂｉ阵是Ｌｉｐｓｃｈｉｔｚ连续的．文献中为了得到非内点连续方法的全局线性收敛性,这一假定是被广泛使用的．本文提出的方法在每一次迭代只须解一个线性方程式组．     

A modified nonmonotone BFGS algorithm for solving smooth nonlinear equations

In this paper, a modified nonmonotone BFGS algorithm is developed for solving a smooth system of nonlinear equations. Different from the existent techniques of nonmonotone line search, the value of an algorithmic parameter controlling the magnitude of nonmonotonicity is updated at each iteration by the known information of the system of nonlinear equations such that the numerical performance of the developed algorithm is improved. Under some suitable assumptions, the global convergence of the algorithm is established for solving a generic nonlinear system of equations. Implementing the algorithm to solve some benchmark test problems, the obtained numerical results demonstrate that it is more effective than some similar algorithms available in the literature.     

一般约束优化基于识别函数的模松弛算法

借助于半罚函数和产生工作集的识别函数以及模松弛SQP算法思想, 本文建立了求解带等式及不等式约束优化的一个新算法. 每次迭代中, 算法的搜索方向由一个简化的二次规划子问题及一个简化的线性方程组产生. 算法在不包含严格互补性的温和条件下具有全局收敛性和超线性收敛性. 最后给出了算法初步的数值试验报告.     

LC1约束规划问题的超线性收敛ODE型信赖域算法

本文对带线性等式约束的LC^1优化问题提出了一个新的ODE型信赖域算法，它在每一次迭代时，不必求解带信赖域界的子问题，仅解一线性方程组而求得试验步。从而可以降低计算的复杂性，提高计算效率，在一定的条件下，文中还证明了该算法是超线性收敛的。     

The selection of the optimal parameter in the modulus-based matrix splitting algorithm for linear complementarity problems

The modulus-based matrix splitting (MMS) algorithm is effective to solve linear complementarity problems (Bai in Numer Linear Algebra Appl 17: 917–933, 2010). This algorithm is parameter dependent, and previous studies mainly focus on giving the convergence interval of the iteration parameter. Yet the specific selection approach of the optimal parameter has not been systematically studied due to the nonlinearity of the algorithm. In this work, we first propose a novel and simple strategy for obtaining the optimal parameter of the MMS algorithm by merely solving two quadratic equations in each iteration. Further, we figure out the interval of optimal parameter which is iteration independent and give a practical choice of optimal parameter to avoid iteration-based computations. Compared with the experimental optimal parameter, the numerical results from three problems, including the Signorini problem of the Laplacian, show the feasibility, effectiveness and efficiency of the proposed strategy.

     

一类变分不等式问题的信赖域算法

基于J.M.Peng研究一类变分不等式问题（简记为VIP）时所提出的价值函数，本文提出了求解强单调的VIP的一个新的信赖域算法。和已有的处理VIP的信赖域方法不同的是：它在每步迭代时，不必求解带信赖域界的子问题，仅解一线性方程组而求得试验步。这样，计算的复杂性一般来说可降低。在通常的假设条件下，文中还证明了算法的整体收敛性。最后，在梯度是半光滑和约束是矩形域的假设下，该算法还是超线性收敛的。     

Generalized monotone line search SQP algorithm for constrained minimax problems

In this article, non-linear minimax problems with general constraints are discussed. By means of solving one quadratic programming an improved direction is yielded and a second-order correction direction can also be at hand via one system of linear equations. So a new algorithm for solving the discussed problems is presented. In connection with a special merit function, the generalized monotone line search is used to yield the step size at each iteration. Under mild conditions, we can ensure global and superlinear convergence. Finally, some numerical experiments are operated to test our algorithm, and the results demonstrate that it is promising.     

A modified SQP algorithm for minimax problems

In this paper, a modified nonmonotone line search SQP algorithm for nonlinear minimax problems is presented. During each iteration of the proposed algorithm, a main search direction is obtained by solving a reduced quadratic program (QP). In order to avoid the Maratos effect, a correction direction is generated by solving the reduced system of linear equations. Under mild conditions, the global and superlinear convergence can be achieved. Finally, some preliminary numerical results are reported.     

Additive preconditioning, eigenspaces, and the inverse iteration

We incorporate our recent preconditioning techniques into the classical inverse power (Rayleigh quotient) iteration for computing matrix eigenvectors. Every loop of this iteration essentially amounts to solving an ill conditioned linear system of equations. Due to our modification we solve a well conditioned linear system instead. We prove that this modification preserves local quadratic convergence, show experimentally that fast global convergence is preserved as well, and yield similar results for higher order inverse iteration, covering the cases of multiple and clustered eigenvalues.     

A new one‐step smoothing newton method for the second‐order cone complementarity problem

In this paper, we present a new one‐step smoothing Newton method for solving the second‐order cone complementarity problem (SOCCP). Based on a new smoothing function, the SOCCP is approximated by a family of parameterized smooth equations. At each iteration, the proposed algorithm only need to solve one system of linear equations and perform only one Armijo‐type line search. The algorithm is proved to be convergent globally and superlinearly without requiring strict complementarity at the SOCCP solution. Moreover, the algorithm has locally quadratic convergence under mild conditions. Numerical experiments demonstrate the feasibility and efficiency of the new algorithm. Copyright © 2010 John Wiley & Sons, Ltd.     

An efficient inexact symmetric Gauss–Seidel based majorized ADMM for high-dimensional convex composite conic programming

In this paper, we propose an inexact multi-block ADMM-type first-order method for solving a class of high-dimensional convex composite conic optimization problems to moderate accuracy. The design of this method combines an inexact 2-block majorized semi-proximal ADMM and the recent advances in the inexact symmetric Gauss–Seidel (sGS) technique for solving a multi-block convex composite quadratic programming whose objective contains a nonsmooth term involving only the first block-variable. One distinctive feature of our proposed method (the sGS-imsPADMM) is that it only needs one cycle of an inexact sGS method, instead of an unknown number of cycles, to solve each of the subproblems involved. With some simple and implementable error tolerance criteria, the cost for solving the subproblems can be greatly reduced, and many steps in the forward sweep of each sGS cycle can often be skipped, which further contributes to the efficiency of the proposed method. Global convergence as well as the iteration complexity in the non-ergodic sense is established. Preliminary numerical experiments on some high-dimensional linear and convex quadratic SDP problems with a large number of linear equality and inequality constraints are also provided. The results show that for the vast majority of the tested problems, the sGS-imsPADMM is 2–3 times faster than the directly extended multi-block ADMM with the aggressive step-length of 1.618, which is currently the benchmark among first-order methods for solving multi-block linear and quadratic SDP problems though its convergence is not guaranteed.