Reinforcement Learning Enhanced Explainer for Graph Neural Networks

Authors: Caihua Shan, Yifei Shen, Yao Zhang, Xiang Li, Dongsheng Li

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on both synthetic and real datasets show that RG-Explainer outperforms state-of-the-art GNN explainers.
Researcher Affiliation Collaboration 1Microsoft Research Asia {caihua.shan,dongsheng.li}@microsoft.com 2The Hong Kong University of Science and Technology yshenaw@connect.ust.hk 3Fudan University yaozhang@fudan.edu.cn 4East China Normal University xiangli@dase.ecnu.edu.cn
Pseudocode Yes Due to the space limitation, we move the pseudocode, implementation details and ablation study to the supplementary materials.
Open Source Code Yes We also attach our codes in the supplementary materials.
Open Datasets Yes We use six datasets, in which four synthetic datasets (BA-shapes, BA-Community, Tree Cycles and Tree-Grid) are used for the node classification task and two datasets (BA-2motifs and Mutagenicity) are used for the graph classificition task.
Dataset Splits No Specifically, we vary the training set sizes from {10%, 30%, 50%, 70%, 90%} and take the remaining instances for testing.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library names with version numbers, needed to replicate the experiment.
Experiment Setup No For fairness, we follow the experimental setup in [17, 12], i.e., the same datasets, trained GNN model and evaluation metrics. Besides, we also utilize the same fine-tuned parameters in [12] for our competitors, GNNExplainer and PGExplainer.