Accelerated Algorithms for Smooth Convex-Concave Minimax Problems
         with O(1/k 2 ) Rate on Squared Gradient Norm

                        TaeHo Yoon 1 Ernest K. Ryu 1

             Abstract                 value for quantifying suboptimality. Moreover, the notion is
   In this work, we study the computational complex-      meaningful for differentiable non-convex-concave minimax
   ity of reducing the squared gradient magnitude        games. Interestingly, very few prior works have analyz ...