Towards Better Generalization of Adaptive Gradient
           Methods

        Yingxue Zhou, Belhal Karimi, Jinxing Yu, Zhiqiang Xu, Ping Li
                 Cognitive Computing Lab
                    Baidu Research
           No.10 Xibeiwang East Road, Beijing 100193, China
           10900 NE 8th St. Bellevue, Washington 98004, USA
  {hustzhouyx, belhal.karimi, jinxingyu18, zhiqiangxu2001, pingli98}@gmail.com

                       Abstract
     Adaptive gradient methods such as AdaGrad, RMS ...