Directional differentiability of the optimal value function in a nonlinear programming problem. (English) Zbl 0546.90088

A parameterized minimization problem is considered in which all the functions involved (the objective function and the constraint ones) are of class \(C^ 2\). Under the assumption that certain multiplier vectors arising in the derivation of second-order necessary conditions for optimality satisfy the ad hoc condition for local optimality based on the augmented Lagrangian, it is shown that the marginal function in the problem, considered as a function of the parameters, has a usual directional derivative. The directional derivatives are expressed by a min-max formula.


90C31 Sensitivity, stability, parametric optimization
49K10 Optimality conditions for free problems in two or more independent variables
49M37 Numerical methods based on nonlinear programming
90C30 Nonlinear programming
Full Text: DOI