Learning and Evaluating Graph Neural Network Explanations based on Counterfactual and Factual Reasoning
Learning and Evaluating Graph Neural Network Explanations based on Counterfactual and Factual Reasoning
复制标题
DOI：
10.1145/3485447.3511948
复制
发表时间：
2022-02
期刊：
影响因子：
--
通讯作者：
Juntao Tan;Shijie Geng;Zuohui Fu;Yingqiang Ge;Shuyuan Xu;Yunqi Li;Yongfeng Zhang
中科院分区：
文献类型：
--
作者：
Juntao Tan;Shijie Geng;Zuohui Fu;Yingqiang Ge;Shuyuan Xu;Yunqi Li;Yongfeng Zhang
结构数据广泛存在于 Web 应用程序中，例如社交媒体中的社交网络、学术网站中的引文网络以及在线论坛中的线程数据。由于拓扑结构复杂，处理和利用此类数据中丰富的信息非常困难。图神经网络（GNN）在学习结构数据表示方面表现出了巨大的优势。然而，深度学习模型的不透明性使得解释和解释 GNN 的预测变得非常重要。同时，评估 GNN 的解释也是一个很大的挑战，因为在很多情况下，无法获得真实的解释。在本文中，我们从因果推理理论中汲取反事实和事实（CF2）推理的见解，来解决可解释的 GNN 中的学习和评估问题。为了生成解释，我们通过基于两个随意视角制定优化问题，提出了一个与模型无关的框架。这将 CF2 与之前仅考虑其中之一的可解释 GNN 区分开来。这项工作的另一个贡献是对 GNN 解释的评估。为了在不需要真实性的情况下定量评估生成的解释，我们设计了基于反事实和事实推理的指标来评估解释的必要性和充分性。实验表明，无论是否有真实的解释，CF2 都能在现实数据集上生成比之前最先进的方法更好的解释。此外，统计分析证明了地面实况评估的表现与我们提出的指标之间的相关性。
Structural data well exists in Web applications, such as social networks in social media, citation networks in academic websites, and threads data in online forums. Due to the complex topology, it is difficult to process and make use of the rich information within such data. Graph Neural Networks (GNNs) have shown great advantages on learning representations for structural data. However, the non-transparency of the deep learning models makes it non-trivial to explain and interpret the predictions made by GNNs. Meanwhile, it is also a big challenge to evaluate the GNN explanations, since in many cases, the ground-truth explanations are unavailable. In this paper, we take insights of Counterfactual and Factual (CF2) reasoning from causal inference theory, to solve both the learning and evaluation problems in explainable GNNs. For generating explanations, we propose a model-agnostic framework by formulating an optimization problem based on both of the two casual perspectives. This distinguishes CF2 from previous explainable GNNs that only consider one of them. Another contribution of the work is the evaluation of GNN explanations. For quantitatively evaluating the generated explanations without the requirement of ground-truth, we design metrics based on Counterfactual and Factual reasoning to evaluate the necessity and sufficiency of the explanations. Experiments show that no matter ground-truth explanations are available or not, CF2 generates better explanations than previous state-of-the-art methods on real-world datasets. Moreover, the statistic analysis justifies the correlation between the performance on ground-truth evaluation and our proposed metrics.