图的可解释性优质论文合集

Explainability：研究图的可解释性。图结构数据本身能够将不同的Entity链接在一起，天然具备较强的解释潜力。早期的图模型主要基于meta-path来做一些可解释性，近年来也有关于诸多GNN做可解释性的文章，值得一读。
小编整理了近期该领域的优质论文，现分享给大家~

1.论文名称：Explainability in Graph Neural Networks: A Taxonomic Survey
链接：https://www.aminer.cn/pub/5fef130291e0113b265a010a
2.论文名称：GNNExplainer: Generating Explanations for Graph Neural Networks
链接：https://www.aminer.cn/pub/5db9293347c8f766461f059b
3.论文名称：Explainability Methods for Graph Convolutional Neural Networks.
链接：https://www.aminer.cn/pub/5da2f8aa3a55ac3402d8c2c5
4.论文名称：Parameterized Explainer for Graph Neural Network
链接：https://www.aminer.cn/pub/5f7fdd328de39f0828397b3f
5.论文名称：XGNN: Towards Model-Level Explanations of Graph Neural Networks
链接：https://www.aminer.cn/pub/5f03f3b611dc83056223206d
6.论文名称：PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks
链接：https://www.aminer.cn/pub/5f7fdd328de39f0828397bdf
7.论文名称：Explanation-based Weakly-supervised Learning of Visual Relations with Graph Networks
链接：https://www.aminer.cn/pub/5eede0b091e0116a23aafa42
8.论文名称：GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media
链接：https://www.aminer.cn/pub/5ea6adfa91e011a546871cbf
9.论文名称：On Explainability of Graph Neural Networks via Subgraph Explorations
链接：https://www.aminer.cn/pub/6025462c91e0118ffcd4a197
10.论文名称：Quantifying Explainers of Graph Neural Networks in Computational Pathology.
链接：https://www.aminer.cn/pub/5fbf836d91e011a96823f10a
11.论文名称：Counterfactual Supporting Facts Extraction for Explainable Medical Record Based Diagnosis with Graph Network
链接：https://www.aminer.cn/pub/60af76f99e795e6b8e55c7b3
12.论文名称：When Comparing to Ground Truth is Wrong - On Evaluating GNN Explanation Methods.
链接：https://www.aminer.cn/pub/6124dd9691e0114cbe7a94b5
13.论文名称：Counterfactual Graphs for Explainable Classification of Brain Networks.
链接：https://www.aminer.cn/pub/60cd529891e011329faa2174
14.论文名称：Explainable Subgraph Reasoning for Forecasting on Temporal Knowledge Graphs
链接：https://www.aminer.cn/pub/6008341b9e795ed227f531e8
15.论文名称：Generative Causal Explanations for Graph Neural Networks
链接：https://www.aminer.cn/pub/6078310791e011f5ecc9dc49
16.论文名称：Improving Molecular Graph Neural Network Explainability with Orthonormalization and Induced Sparsity
链接：https://www.aminer.cn/pub/609baf6491e0113c3c769369
17.论文名称：Explainable Automated Graph Representation Learning with Hyperparameter Importance.
链接：https://www.aminer.cn/pub/60f16ace91e011963c8d3ca6
18.论文名称：Higher-Order Explanations of Graph Neural Networks via Relevant Walks.
链接：https://www.aminer.cn/pub/6153e3575244ab9dcb3d31fc
19.论文名称：HENIN: Learning Heterogeneous Neural Interaction Networks for Explainable Cyberbullying Detection on Social Media
链接：https://www.aminer.cn/pub/5f7fe6d80205f07f6897320b
更多优质论文，尽在AMiner，主页添加关键词，系统智能推荐最新优质论文~
AMiner平台链接：https://www.aminer.cn/?f=cs