Incorporating Retrieval-based Causal Learning with Information Bottlenecks for Interpretable Graph Neural Networks
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911772779216896 |
|---|---|
| author | Rao, Jiahua Xie, Jiancong Lin, Hanjing Zheng, Shuangjia Wang, Zhen Yang, Yuedong |
| author_facet | Rao, Jiahua Xie, Jiancong Lin, Hanjing Zheng, Shuangjia Wang, Zhen Yang, Yuedong |
| contents | Graph Neural Networks (GNNs) have gained considerable traction for their capability to effectively process topological data, yet their interpretability remains a critical concern. Current interpretation methods are dominated by post-hoc explanations to provide a transparent and intuitive understanding of GNNs. However, they have limited performance in interpreting complicated subgraphs and can't utilize the explanation to advance GNN predictions. On the other hand, transparent GNN models are proposed to capture critical subgraphs. While such methods could improve GNN predictions, they usually don't perform well on explanations. Thus, it is desired for a new strategy to better couple GNN explanation and prediction. In this study, we have developed a novel interpretable causal GNN framework that incorporates retrieval-based causal learning with Graph Information Bottleneck (GIB) theory. The framework could semi-parametrically retrieve crucial subgraphs detected by GIB and compress the explanatory subgraphs via a causal module. The framework was demonstrated to consistently outperform state-of-the-art methods, and to achieve 32.71\% higher precision on real-world explanation scenarios with diverse explanation types. More importantly, the learned explanations were shown able to also improve GNN prediction performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_04710 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Incorporating Retrieval-based Causal Learning with Information Bottlenecks for Interpretable Graph Neural Networks Rao, Jiahua Xie, Jiancong Lin, Hanjing Zheng, Shuangjia Wang, Zhen Yang, Yuedong Machine Learning Graph Neural Networks (GNNs) have gained considerable traction for their capability to effectively process topological data, yet their interpretability remains a critical concern. Current interpretation methods are dominated by post-hoc explanations to provide a transparent and intuitive understanding of GNNs. However, they have limited performance in interpreting complicated subgraphs and can't utilize the explanation to advance GNN predictions. On the other hand, transparent GNN models are proposed to capture critical subgraphs. While such methods could improve GNN predictions, they usually don't perform well on explanations. Thus, it is desired for a new strategy to better couple GNN explanation and prediction. In this study, we have developed a novel interpretable causal GNN framework that incorporates retrieval-based causal learning with Graph Information Bottleneck (GIB) theory. The framework could semi-parametrically retrieve crucial subgraphs detected by GIB and compress the explanatory subgraphs via a causal module. The framework was demonstrated to consistently outperform state-of-the-art methods, and to achieve 32.71\% higher precision on real-world explanation scenarios with diverse explanation types. More importantly, the learned explanations were shown able to also improve GNN prediction performance. |
| title | Incorporating Retrieval-based Causal Learning with Information Bottlenecks for Interpretable Graph Neural Networks |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.04710 |