In the problem of structured prediction with graph representation learning (GRL for short), the hypothesis returned by the algorithm maps the set of features in the \emph{receptive field} of the targeted vertex to its label. To understand the learnability of those algorithms, we introduce a weaker form of uniform stability termed \emph{multi-fidelity stability} and give learning guarantees for weakly dependent graphs. We testify that ~\citet{london2016stability}'s claim on the generalization of a single sample holds for GRL when the receptive field is sparse. In addition, we study the stability induced bound for two popular algorithms: \textbf{(1)} Stochastic gradient descent under convex and non-convex landscape. In this example, we provide non-asymptotic bounds that highly depend on the sparsity of the receptive field constructed by the algorithm. \textbf{(2)} The constrained regression problem on a 1-layer linear equivariant GNN. In this example, we present lower bounds for the discrepancy between the two types of stability, which justified the multi-fidelity design.
Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networks
School of Computing and Information Sciences, Florida International University(佛罗里达国际大学计算与信息科学学院);NEC Laboratories America(美国NEC实验室);School of Computing and Augmented Intelligence, Arizona State University(亚利桑那州立大学计算与增强智能学院)
Perturb More, Trap More: Understanding Behaviors of Graph Neural Networks
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(中国科学院深圳先进技术研究院);Joint Engineering Research Center for Health Big Data Intelligent Analysis Technology(健康大数据智能分析技术联合工程研究中心)
Department of Aerospace and Mechanical Engineering, University of Southern California(航空航天与机械工程系,南加州大学);School of Mathematics, University of Birmingham(数学学院,伯明翰大学);Computing and Mathematical Sciences, California Institute of Technology(计算与数学科学系,加州理工学院)