Formal Verification of Graph Convolutional Networks with Uncertain Node Features and Uncertain Graph Structure
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910912188776448 |
|---|---|
| author | Ladner, Tobias Eichelbeck, Michael Althoff, Matthias |
| author_facet | Ladner, Tobias Eichelbeck, Michael Althoff, Matthias |
| contents | Graph neural networks are becoming increasingly popular in the field of machine learning due to their unique ability to process data structured in graphs. They have also been applied in safety-critical environments where perturbations inherently occur. However, these perturbations require us to formally verify neural networks before their deployment in safety-critical environments as neural networks are prone to adversarial attacks. While there exists research on the formal verification of neural networks, there is no work verifying the robustness of generic graph convolutional network architectures with uncertainty in the node features and in the graph structure over multiple message-passing steps. This work addresses this research gap by explicitly preserving the non-convex dependencies of all elements in the underlying computations through reachability analysis with (matrix) polynomial zonotopes. We demonstrate our approach on three popular benchmark datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_15065 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Formal Verification of Graph Convolutional Networks with Uncertain Node Features and Uncertain Graph Structure Ladner, Tobias Eichelbeck, Michael Althoff, Matthias Machine Learning Artificial Intelligence Graph neural networks are becoming increasingly popular in the field of machine learning due to their unique ability to process data structured in graphs. They have also been applied in safety-critical environments where perturbations inherently occur. However, these perturbations require us to formally verify neural networks before their deployment in safety-critical environments as neural networks are prone to adversarial attacks. While there exists research on the formal verification of neural networks, there is no work verifying the robustness of generic graph convolutional network architectures with uncertainty in the node features and in the graph structure over multiple message-passing steps. This work addresses this research gap by explicitly preserving the non-convex dependencies of all elements in the underlying computations through reachability analysis with (matrix) polynomial zonotopes. We demonstrate our approach on three popular benchmark datasets. |
| title | Formal Verification of Graph Convolutional Networks with Uncertain Node Features and Uncertain Graph Structure |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2404.15065 |