MIQCQP reformulation of the ReLU neural networks Lipschitz constant estimation problem
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909090825895936 |
|---|---|
| author | Sbihi, Mohammed Jan, Sophie Couellan, Nicolas |
| author_facet | Sbihi, Mohammed Jan, Sophie Couellan, Nicolas |
| contents | It is well established that to ensure or certify the robustness of a neural network, its Lipschitz constant plays a prominent role. However, its calculation is NP-hard. In this note, by taking into account activation regions at each layer as new constraints, we propose new quadratically constrained MIP formulations for the neural network Lipschitz estimation problem. The solutions of these problems give lower bounds and upper bounds of the Lipschitz constant and we detail conditions when they coincide with the exact Lipschitz constant. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01199 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MIQCQP reformulation of the ReLU neural networks Lipschitz constant estimation problem Sbihi, Mohammed Jan, Sophie Couellan, Nicolas Optimization and Control Machine Learning It is well established that to ensure or certify the robustness of a neural network, its Lipschitz constant plays a prominent role. However, its calculation is NP-hard. In this note, by taking into account activation regions at each layer as new constraints, we propose new quadratically constrained MIP formulations for the neural network Lipschitz estimation problem. The solutions of these problems give lower bounds and upper bounds of the Lipschitz constant and we detail conditions when they coincide with the exact Lipschitz constant. |
| title | MIQCQP reformulation of the ReLU neural networks Lipschitz constant estimation problem |
| topic | Optimization and Control Machine Learning |
| url | https://arxiv.org/abs/2402.01199 |