Stability of saddle points via explicit coderivatives of pointwise subdifferentials (Q526389)

The authors investigate the minimizers of convex composed optimization problems defined on Hilbert spaces, providing stability criteria for the corresponding saddle points by means of metric regularity of attached infinite-dimensional set-valued mappings. As the authors point out, ``a main ingredient [for this study] is an explicit pointwise characterization of the regular coderivative of the subdifferential of convex integral functionals.'' Applications to stability properties for parameter identification problems for an elliptic partial differential equation with non-differentiable data fitting terms are given afterwards, followed by an appendix dedicated to a technical result.