Towards More Practical Adversarial Attacks on
         Graph Neural Networks

       Jiaqi Ma           Shuangrui Ding           Qiaozhu Mei
     jiaqima@umich.edu        markding@umich.edu          qmei@umich.edu

                        Abstract
      We study the black-box attacks on graph neural networks (GNNs) under a novel
      and realistic constraint: attackers have access to only a subset of nodes in the
      network, and they can only attack a small number of them. A node selec ...