Formal Verification of Graph Convolutional Networks with Uncertain Node Features and Uncertain Graph Structure

no code implementations • 23 Apr 2024 • Tobias Ladner, Michael Eichelbeck, Matthias Althoff

Graph neural networks are becoming increasingly popular in the field of machine learning due to their unique ability to process data structured in graphs.

Set-Based Training for Neural Network Verification

no code implementations • 26 Jan 2024 • Lukas Koller, Tobias Ladner, Matthias Althoff

Neural networks are vulnerable to adversarial attacks, i. e., small input perturbations can significantly affect the outputs of a neural network.

Fully Automatic Neural Network Reduction for Formal Verification

no code implementations • 3 May 2023 • Tobias Ladner, Matthias Althoff

Our evaluation shows that our approach can reduce the number of neurons to a fraction of the original number of neurons with minor outer-approximation and thus reduce the verification time to a similar degree.