This paper establishes the iteration-complexity of proximal bundle methods for solving hybrid (i.e., a blend of smooth and nonsmooth) weakly convex composite optimization (HWC-CO) problems. This is done in a unified manner by considering a proximal bundle framework (PBF) based on a generic bundle update scheme which includes various well-known bundle update schemes. In contrast to other well-known stationary conditions in the context of HWC-CO, PBF uses a new stationarity measure which is easily verifiable and, at the same time, implies any of the former ones.
Article
View Proximal bundle methods for hybrid weakly convex composite optimization problems