SOL: Sampling-based Optimal Linear bounding of arbitrary scalar functions
–Neural Information Processing Systems
Finding tight linear bounds for activation functions in neural networks is an essential part of several state of the art neural network robustness certification tools.
Neural Information Processing Systems
Feb-13-2026, 05:57:34 GMT
- Country:
- North America > United States
- California (0.14)
- Georgia > Fulton County
- Atlanta (0.04)
- Europe
- Portugal > Porto
- Porto (0.04)
- Germany
- Bavaria > Upper Bavaria
- Munich (0.04)
- Baden-Württemberg > Karlsruhe Region
- Heidelberg (0.04)
- Bavaria > Upper Bavaria
- Portugal > Porto
- North America > United States
- Technology: