Li, Y.; Zhou, J.; Zheng, B.; Shafiabady, N.; Chen, F. Reliable and Faithful Generative Explainers for Graph Neural Networks. Preprints2024, 2024101718. https://doi.org/10.20944/preprints202410.1718.v1
APA Style
Li, Y., Zhou, J., Zheng, B., Shafiabady, N., & Chen, F. (2024). Reliable and Faithful Generative Explainers for Graph Neural Networks. Preprints. https://doi.org/10.20944/preprints202410.1718.v1
Chicago/Turabian Style
Li, Y., Niusha Shafiabady and Fang Chen. 2024 "Reliable and Faithful Generative Explainers for Graph Neural Networks" Preprints. https://doi.org/10.20944/preprints202410.1718.v1
Abstract
Graph neural networks (GNNs) have been effectively implemented in a variety of real-world applications, while their underlying work mechanisms remain a mystery. To unveil this mystery and advocate trustworthy decision-making, many GNN explainers have been proposed. However, existing explainers often face significant challenges, such as: 1) explanations being tied to specific instances; 2) limited generalizability to unseen graphs; 3) potential generation of invalid graph structures; and 4) restrictions to particular tasks (e.g., node classification, graph classification). To address these challenges, we propose a novel explainer, GAN-GNNExplainer, which employs a generator to produce explanations and a discriminator to oversee the generation process, enhancing the reliability of the outputs. Despite its advantages, GAN-GNNExplainer still struggles with generating faithful explanations and underperforms on real-world datasets. To overcome these shortcomings, we introduce ACGAN-GNNExplainer, an improved approach that improves upon GAN-GNNExplainer by using a more robust discriminator that consistently monitors the generation process, thereby producing explanations that are both reliable and faithful. Extensive experiments on both synthetic and real-world graph datasets demonstrate the superiority of our proposed methods over existing GNN explainers.
Computer Science and Mathematics, Computer Science
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
