Stability and Generalization of Stochastic Gradient Methods for Minimax
                    Problems

              Yunwen Lei * 1 Zhenhuan Yang * 2 Tianbao Yang 3 Yiming Ying 2

             Abstract                 ing (Goodfellow et al., 2014), robust optimization (Chen
   Many machine learning problems can be formu-        et al., 2017; Namkoong & Duchi, 2017), reinforcement
   lated as minimax problems such as Generative        learning (Dai et al., 2018; Du et al., 2017) and AUC max-
   Ad ...