Explaining Graph Neural Networks for Node Similarity on Graphs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917480074575872 |
|---|---|
| author | Daza, Daniel Chu, Cuong Xuan Tran, Trung-Kien Stepanova, Daria Cochez, Michael Groth, Paul |
| author_facet | Daza, Daniel Chu, Cuong Xuan Tran, Trung-Kien Stepanova, Daria Cochez, Michael Groth, Paul |
| contents | Similarity search is a fundamental task for exploiting information in various applications dealing with graph data, such as citation networks or knowledge graphs. While this task has been intensively approached from heuristics to graph embeddings and graph neural networks (GNNs), providing explanations for similarity has received less attention. In this work we are concerned with explainable similarity search over graphs, by investigating how GNN-based methods for computing node similarities can be augmented with explanations. Specifically, we evaluate the performance of two prominent approaches towards explanations in GNNs, based on the concepts of mutual information (MI), and gradient-based explanations (GB). We discuss their suitability and empirically validate the properties of their explanations over different popular graph benchmarks. We find that unlike MI explanations, gradient-based explanations have three desirable properties. First, they are actionable: selecting inputs depending on them results in predictable changes in similarity scores. Second, they are consistent: the effect of selecting certain inputs overlaps very little with the effect of discarding them. Third, they can be pruned significantly to obtain sparse explanations that retain the effect on similarity scores. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_07639 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Explaining Graph Neural Networks for Node Similarity on Graphs Daza, Daniel Chu, Cuong Xuan Tran, Trung-Kien Stepanova, Daria Cochez, Michael Groth, Paul Machine Learning Artificial Intelligence Similarity search is a fundamental task for exploiting information in various applications dealing with graph data, such as citation networks or knowledge graphs. While this task has been intensively approached from heuristics to graph embeddings and graph neural networks (GNNs), providing explanations for similarity has received less attention. In this work we are concerned with explainable similarity search over graphs, by investigating how GNN-based methods for computing node similarities can be augmented with explanations. Specifically, we evaluate the performance of two prominent approaches towards explanations in GNNs, based on the concepts of mutual information (MI), and gradient-based explanations (GB). We discuss their suitability and empirically validate the properties of their explanations over different popular graph benchmarks. We find that unlike MI explanations, gradient-based explanations have three desirable properties. First, they are actionable: selecting inputs depending on them results in predictable changes in similarity scores. Second, they are consistent: the effect of selecting certain inputs overlaps very little with the effect of discarding them. Third, they can be pruned significantly to obtain sparse explanations that retain the effect on similarity scores. |
| title | Explaining Graph Neural Networks for Node Similarity on Graphs |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2407.07639 |