This paper considers the nonlinear symmetric conic programming (NSCP) problems. Firstly, a type of strong sufficient optimality condition for NSCP problems in terms of a linear-quadratic term is introduced. Then, a sufficient condition of the nonsingularity of Clarke’s generalized Jacobian of the Karush–Kuhn–Tucker (KKT) system is demonstrated. At last, as an application, this property is used to obtain the local convergence properties of a sequential quadratic programming- (SQP-) type method.
This paper studies the nonlinear programming problem with quaternion self-conjugate negative semidefinite matrix constraints(QNSDP). A theoretical framework for QNSDP is constructed, and an efficient numerical algorithm is designed. First, a projection operator on the space of quaternion self-conjugate matrices is defined, and constraint qualifications such as the Robinson constraint qualification, the constraint nondegeneracy condition, and the strict complementarity condition are provided. It is proved that the Lagrange multiplier set is nonempty and bounded if and only if the Robinson constraint qualification holds, among other results. Second, the KKT conditions and second-order optimality conditions are established, and an explicit expression for the critical cone is given, then SQP-type methods for solving QNSDP is proposed. Under the different assumptions, the convergence of the algorithm is proved. Finally, numerical experiments verify the effectiveness of the algorithm.
25 April 2022 in Optimization Methods and Software
In this paper, we discuss the isotonicity of the proximity operator in Hilbert quasi-lattices endowed with different Lorentz cones. The extended Lorentz cone is first defined by the Minkowski functionals of some subsets. We then establish some sufficient conditions for the isotonicity of the proximity operator concerning one order and two mutually dual orders induced by Lorentz cones, respectively. Similarly, the cases of the extended Lorentz cones and other ordered inequality properties of the proximity operator are analysed. By adopting these characterizations, some solvability and iterative algorithm theorems for the stochastic optimization problem are established by different order approaches. For solvability, the gradient of the mappings does not need to be continuous, and the solutions are optimal with respect to the orders. In the stochastic proximal algorithms, the mappings satisfy inequality conditions just for comparable elements, but the convergence direction and convergence rate are more optimal.
1 January 2015 in Journal of Industrial and Management Optimization
Since the monotonicity of the best approximant is crucial to establish partial ordering methods, in this paper, we, respectively, characterize the best approximants in Banach function spaces and Lorentz spaces , in which we especially focus on the monotonicity characterizations. We first study monotonicity characterizations of the metric projection operator onto sublattices in general Banach function spaces by the property . The sufficient and necessary conditions for monotonicity of the metric projection onto cones and sublattices are then, respectively, established in . The Lorentz spaces are also shown to be reflexive under the condition , which is the basis for the existence of the best approximant. As applications, by establishing the partial ordering methods based on the obtained monotonicity characterizations, the solvability and approximation theorems for best proximity points are deduced without imposing any contractive and compact conditions in . Our results extend and improve many previous results in the field of the approximation and partial ordering theory.