Stability in the presence of degeneracy and error estimation

Given an approximation to a local minimizer to a nonlinear optimization problem and to associated multipliers, we obtain a tight error estimate in terms of the violation of the first-order conditions. Our results apply to degenerate optimization problems where independence of the active constraint gradients and strict complementarity can be violated.

[1]  Stephen J. Wright.  Superlinear Convergence of a Stabilized SQP Method to a Degenerate Solution , 1998, Comput. Optim. Appl..

[2]  Stephen M. Robinson,et al.  Strongly Regular Generalized Equations , 1980, Math. Oper. Res..

[3]  O. Mangasarian,et al.  The Fritz John Necessary Optimality Conditions in the Presence of Equality and Inequality Constraints , 1967 .

[4]  D. Klatte,et al.  On the Lipschitz behavior of optimal solutions in parametric problems of quadratic optimization and linear complementarity , 1985 .

[5]  D. Klatte,et al.  Lipschitz continuity of infima and optimal solutions in parametric optimization: the polyhedral case , 1987 .

[6]  W. Hager,et al.  Lipschitzian stability in nonlinear control and optimization , 1993 .

[7]  D. Klatte.  Nonlinear Optimization Problems under Data Perturbations , 1992 .

[8]  Asen L. Dontchev,et al.  Characterizations of Lipschitz Stability in Optimization , 1995 .

[9]  R. Wets,et al.  A Lipschitzian characterization of convex polyhedra , 1969 .

[10]  S. M. Robinson.  Some continuity properties of polyhedral multifunctions , 1981 .

[11]  A. Hoffman.  On approximate solutions of systems of linear inequalities , 1952 .

[12]  Francisco Facchinei,et al.  On the Accurate Identification of Active Constraints , 1998, SIAM J. Optim..

[13]  S. M. Robinson.  Stability Theory for Systems of Inequalities. Part I: Linear Systems , 1975 .

[14]  S. M. Robinson.  Generalized equations and their solutions, part II: Applications to nonlinear programming , 1982 .