In this paper, we extend the results for approximation semigroups for general resolvent maps including various resolvents of maps on a general convex geodesic metric space. For our study, we introduce the notion of (g...
详细信息
In this paper, we extend the results for approximation semigroups for general resolvent maps including various resolvents of maps on a general convex geodesic metric space. For our study, we introduce the notion of (general) resolvent maps which is a generalization of the resolvent maps in Lawson (J Lie Theory 33, 361-376, 2023) and then we prove several useful properties for the resolvent map and construct the approximation semigroups for resolvent maps. We also study the convergence of a proximalpoint like algorithm for the general resolvent map.
We present two new error bounds for optimization problems over a convex set whose objective function f is either semianalytic or gamma-strictly convex, with gamma greater than or equal to 1. We then apply these error ...
详细信息
We present two new error bounds for optimization problems over a convex set whose objective function f is either semianalytic or gamma-strictly convex, with gamma greater than or equal to 1. We then apply these error bounds to analyze the rate of convergence of a wide class of iterative descent algorithms for the aforementioned optimization problem. Our analysis shows that the function sequence {f(x(k))} converges at least at the sublinear rate of k(-epsilon) for some positive constant epsilon, where k is the iteration index. Moreover, the distances from the iterate sequence {x(k)} to the set of stationary points of the optimization problem converge to zero at least sublinearly.
In this paper, we present a logarithmic-quadratic proximal (LQP) type prediction-correction methods for solving constrained variational inequalities VI(S,f), where S is a convex set with linear constraints. The comput...
详细信息
In this paper, we present a logarithmic-quadratic proximal (LQP) type prediction-correction methods for solving constrained variational inequalities VI(S,f), where S is a convex set with linear constraints. The computational load in each iteration is quite tiny. However, the number of iterations is significantly dependent on a parameter which balances the primal and dual variables. We then propose a self-adaptive prediction-correction method that adjusts the scalar parameter automatically. Under certain conditions, the global convergence of the proposed method is established. In order to demonstrate the efficiency of the proposed method, we provide numerical results for a convex nonlinear programming and traffic equilibrium problems. (c) 2006 Elsevier Inc. All rights reserved.
We consider a multiple-block separable convex programming problem, where the objective function is the sum of m individual convex functions without overlapping variables, and the constraints are linear, aside from sid...
详细信息
We consider a multiple-block separable convex programming problem, where the objective function is the sum of m individual convex functions without overlapping variables, and the constraints are linear, aside from side constraints. Based on the combination of the classical Gauss-Seidel and the Jacobian decompositions of the augmented Lagrangian function, we propose a partially parallel splitting method, which differs from existing augmented Lagrangian based splitting methods in the sense that such an approach simplifies the iterative scheme significantly by removing the potentially expensive correction step. Furthermore, a relaxation step, whose computational cost is negligible, can be incorporated into the proposed method to improve its practical performance. Theoretically, we establish global convergence of the new method in the framework of proximal point algorithm and worst-case nonasymptotic O(1/t) convergence rate results in both ergodic and nonergodic senses, where t counts the iteration. The efficiency of the proposed method is further demonstrated through numerical results on robust PCA, i.e., factorizing from incomplete information of an unknown matrix into its low-rank and sparse components, with both synthetic and real data of extracting the background from a corrupted surveillance video.
Primal-dual hybrid gradient (PDHG) method is a canonical and popular prototype for solving saddle point problem (SPP). However, the nonlinear coupling term in SPP excludes the application of PDHG on far-reaching real-...
详细信息
Primal-dual hybrid gradient (PDHG) method is a canonical and popular prototype for solving saddle point problem (SPP). However, the nonlinear coupling term in SPP excludes the application of PDHG on far-reaching real-world problems. In this paper, following the seminal work by Valkonen (Inverse Problems 30, 2014), we devise a variant iterative scheme for solving SPP with nonlinear function by exerting an alternative extrapolation procedure. The novel iterative scheme falls exactly into the proximal point algorithmic framework without any residuals, which indicates that the associated inclusion problem is "nearer" to the KKT mapping induced by SPP. Under the metrically regular assumption on KKT mapping, we simplify the local convergence of the proposed method on contractive perspective. Numerical simulations on a PDE-constrained nonlinear inverse problem demonstrate the compelling performance of the proposed method.
In this paper, a solution to the inclusion problem for an infinite family of monotone operators in Hadamard spaces is approximated. Strong convergence and Delta-convergence to a common zero of an infinite family of mo...
详细信息
In this paper, a solution to the inclusion problem for an infinite family of monotone operators in Hadamard spaces is approximated. Strong convergence and Delta-convergence to a common zero of an infinite family of monotone operators are established. To support these results, some applications in convex optimization and fixed point theory are also presented.
In the alternating directions method, the relaxation factor gamma is an element of (0, root 5+1/2) by Glowinski is useful in practical computations for structured variational inequalities. This paper points out that t...
详细信息
In the alternating directions method, the relaxation factor gamma is an element of (0, root 5+1/2) by Glowinski is useful in practical computations for structured variational inequalities. This paper points out that the same restriction region of the relaxation factor is also valid in the proximal alternating directions method.
This paper considers the problem of finding a zero of the sum of a single-valued Lipschitz continuous mapping A and a maximal monotone mapping B in a closed convex set C. We first give some projection-type methods and...
详细信息
This paper considers the problem of finding a zero of the sum of a single-valued Lipschitz continuous mapping A and a maximal monotone mapping B in a closed convex set C. We first give some projection-type methods and extend a modified projection method proposed by Solodov and Tseng for the special case of B = N(c) to this problem, then we give a refinement of Tseng's method that replaces P(c) by P(ck). Finally, convergence of these methods is established. (C) 2010 Elsevier B.V. All rights reserved.
In this paper, we propose a proximal parallel decomposition algorithm for solving the optimization problems where the objective function is the sum of m separable functions (i.e., they have no crossed variables), and ...
详细信息
In this paper, we propose a proximal parallel decomposition algorithm for solving the optimization problems where the objective function is the sum of m separable functions (i.e., they have no crossed variables), and the constraint set is the intersection of Cartesian products of some simple sets and a linear manifold. The m subproblems are solved simultaneously per iterations, which are sum of the decomposed subproblems of the augmented Lagrange function and a quadratic term. Hence our algorithm is named as the 'proximal parallel splitting method'. We prove the global convergence of the proposed algorithm under some mild conditions that the underlying functions are convex and the solution set is nonempty. To make the subproblems easier, some linearized versions of the proposed algorithm are also presented, together with their global convergence analysis. Finally, some preliminary numerical results are reported to support the efficiency of the new algorithms. (C) 2013 Elsevier B.V. All rights reserved.
The Fantope-constrained sparse principal subspace estimation problem is initially proposed Vu et al. (Vu et al., 2013). This paper investigates a semismooth Newton based proximalpoint (P PASSN ) algorithm for solving...
详细信息
The Fantope-constrained sparse principal subspace estimation problem is initially proposed Vu et al. (Vu et al., 2013). This paper investigates a semismooth Newton based proximalpoint (P PASSN ) algorithm for solving the equivalent form of this problem, where a semismooth Newton (S SN ) method is utilized to optimize the inner problems involved in the P PASSN algorithm. Under standard conditions, the P PASSN algorithm is proven to achieve global convergence and asymptotic superlinear convergence rate. Computationally, we derive nontrivial expressions the Fantope projection and its generalized Jacobian, which are key ingredients for the P PASSN algorithm. Some numerical results on synthetic and real data sets are presented to illustrate the effectiveness of the proposed P PASSN algorithm for large-scale problems and superiority over the alternating direction method of multipliers (ADMM).
暂无评论