GNN Explainers

Graph Neural Network (GNN) explainers aim to make the "black box" predictions of GNNs more understandable by identifying the crucial subgraphs or features driving model decisions. Current research focuses on developing more robust and accurate explainers, addressing challenges like susceptibility to adversarial attacks and the lack of reliable ground truth for evaluation. This is achieved through various approaches, including causal inference models, Shapley value methods, and generative models, with a growing emphasis on human-centered design and interactive explanation frameworks. Improved GNN explainability is crucial for building trust in GNN applications across diverse fields, from drug discovery to social network analysis, by providing insights into model behavior and facilitating collaboration between AI experts and domain specialists.

Papers

July 12, 2024

Graph Neural Network Causal Explanation via Neural Causal Models
Arman Behnam, Binghui Wang
Graph Neural Network Causal Structure Structural Causal Model Causal Explanation Neural Causal Model GNN Explainers

June 10, 2024

Explainable Graph Neural Networks Under Fire
Zhong Li, Simon Geisler, Yuhang Wang, Stephan Günnemann, Matthijs van Leeuwen
Graph Neural Network GNN Explanation Fire Occurrence GNN Explainers

May 11, 2024

Design Requirements for Human-Centered Graph Neural Network Explanations
Pantea Habibi, Peyman Baghershahi, Sourav Medya, Debaleena Chattopadhyay
Graph Neural Network Graph Learning GNN Explanation GNN Explainers Design Consideration Human Understandable Explanation Graph Neural Network Explanation

January 26, 2024

GOAt: Explaining Graph Neural Networks via Graph Output Attribution
Shengyao Lu, Keith G. Mills, Jiao He, Bang Liu, Di Niu
Graph Neural Network Graph Feature Electric SHEEP GNN Explainers Graph Neural Network Attribution

January 9, 2024

GNNShap: Scalable and Accurate GNN Explanation using Shapley Values
Selahattin Akkas, Ariful Azad
Gene Level GNN Shapley Value GNN Explanation GNN Explainers

September 29, 2023

ACGAN-GNNExplainer: Auxiliary Conditional Generative Explainer for Graph Neural Networks
Yiqiao Li, Jianlong Zhou, Yifei Dong, Niusha Shafiabady, Fang Chen
Graph Neural Network GNN Explanation Auxiliary Classifier GNN Explainers Generative Explanation

May 21, 2023

Self-Explainable Graph Neural Networks for Link Prediction
Huaisheng Zhu, Dongsheng Luo, Xianfeng Tang, Junjie Xu, Hui Liu, Suhang Wang
Graph Neural Network Link Prediction Post Hoc Explanation GNN Explainers Self Explainable Graph Neural Network

March 4, 2023

DAG Matters! GFlowNets Enhanced Explainer For Graph Neural Networks
Wenqian Li, Yinchuan Li, Zhigang Li, Jianye Hao, Yan Pang
Graph Neural Network Energy Based GFlowNets GNN Explainers Generative Grammar Dag DLT

November 3, 2022

INGREX: An Interactive Explanation Framework for Graph Neural Networks
Tien-Cuong Bui, Van-Duc Le, Wen-Syan Li, Sang Kyun Cha
Graph Neural Network Model Understanding GNN Explainers Explanation Algorithm Explanation Framework

October 27, 2022

Explaining the Explainers in Graph Neural Networks: a Comparative Study
Antonio Longa, Steve Azzolin, Gabriele Santin, Giulia Cencetti, Pietro Liò, Bruno Lepri, Andrea Passerini
Graph Neural Network Comparative Study Evaluation Metric GNN Architecture GNN Explainers Node Classification Datasets

August 19, 2022

Evaluating Explainability for Graph Neural Networks
Chirag Agarwal, Owen Queen, Himabindu Lakkaraju, Marinka Zitnik
Graph Neural Network Real World Graph Synthetic Graph GNN Explanation Graph Explainability GNN Explainers

February 17, 2022

Learning and Evaluating Graph Neural Network Explanations based on Counterfactual and Factual Reasoning
Juntao Tan, Shijie Geng, Zuohui Fu, Yingqiang Ge, Shuyuan Xu, Yunqi Li, Yongfeng Zhang
LeArning Abstract Graph Neural Network Ground Truth GNN Explainers Graph Neural Network Explanation