From Dual Tracking to Clipping: Provably Faster Distributionally Robust Multi-Objective Optimization
This work introduces distributionally robust multi-objective optimization (DR-MOO), which minimizes multiple objectives under their respective worst-case distributions, and proposes Pareto-type solution concepts for DR-MOO and develops multi-gradient descent algorithms (MGDA) with provable guarantees.
Yu-Feng Yang, Fangning Zhuo, Zi-Yi Chen et al.
· 0 citations