Parameterized Explainer for Graph Neural Network
Parameterized Explainer for Graph Neural Network
复制标题
DOI：
--
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Dongsheng Luo;Wei Cheng;Dongkuan Xu;Wenchao Yu;Bo Zong;Haifeng Chen;Xiang Zhang
中科院分区：
文献类型：
--
作者：
Dongsheng Luo;Wei Cheng;Dongkuan Xu;Wenchao Yu;Bo Zong;Haifeng Chen;Xiang Zhang
尽管最近在图神经网络（GNNs）方面取得了进展，但解释GNNs做出的预测仍然是一个具有挑战性的开放问题。引导方法独立地解决局部解释（即，重要的子图结构和节点特征）来解释为什么GNN模型对单个实例（例如节点或图）进行预测。因此，生成的解释是为每个实例精心定制的。独立解释每个实例的独特解释不足以提供对学习的GNN模型的全局理解，导致缺乏可推广性并阻碍其在归纳环境中使用。此外，由于它是为解释单个实例而设计的，因此自然地解释一组实例是具有挑战性的（例如，给定类的图形）。在这项研究中，我们解决了这些关键挑战，并提出了PGExplainer，一个用于GNN的参数化解释器。PGExplainer采用深度神经网络来参数化解释的生成过程，这使得PGExplainer能够自然地集体解释多个实例。与现有的工作相比，PGExplainer具有更好的泛化能力，可以很容易地在归纳设置中使用。在合成数据集和真实数据集上的实验显示出高度竞争性的性能，在解释图分类方面，AUC相对于领先基线的相对改善高达24.7%。
Despite recent progress in Graph Neural Networks (GNNs), explaining predictions made by GNNs remains a challenging open problem. The leading method independently addresses the local explanations (i.e., important subgraph structure and node features) to interpret why a GNN model makes the prediction for a single instance, e.g. a node or a graph. As a result, the explanation generated is painstakingly customized for each instance. The unique explanation interpreting each instance independently is not sufficient to provide a global understanding of the learned GNN model, leading to a lack of generalizability and hindering it from being used in the inductive setting. Besides, as it is designed for explaining a single instance, it is challenging to explain a set of instances naturally (e.g., graphs of a given class). In this study, we address these key challenges and propose PGExplainer, a parameterized explainer for GNNs. PGExplainer adopts a deep neural network to parameterize the generation process of explanations, which enables PGExplainer a natural approach to explaining multiple instances collectively. Compared to the existing work, PGExplainer has better generalization ability and can be utilized in an inductive setting easily. Experiments on both synthetic and real-life datasets show highly competitive performance with up to 24.7\% relative improvement in AUC on explaining graph classification over the leading baseline.