10 papers
Differential Zonotopes for Verifying Global Robustness of DNNs
Anagha Athavale, Samuel Teuber, Matteo Maffei +3
The robustness of deep neural networks (DNNs) is critical in security-sensitive applications, where small input perturbations should not alter model predictions. This property is c…
Compositional Neural-Cyber-Physical System Verification in the Interactive Theorem Prover of Your Choice
Matthew L. Daggitt, Ekaterina Komendantskaya, Alistair Sirman +4
Formal verification of neuro-symbolic cyber-physical systems, such as drones, medical devices and robots, is complicated. Neural components must be trained to be optimal with respe…
Encrypted Neural Networks without Overflows
Philipp Kern, Lorenzo Rovida, Samuel Teuber +3
The popular Cheon-Kim-Kim-Song (CKKS) scheme enables efficient private inference in neural networks by evaluating them on encrypted data. Since CKKS only supports addition, multipl…
Heterogeneous Dynamic Logic: Provability Modulo Program Theories
Samuel Teuber, Mattias Ulbrich, André Platzer +1
Formally specifying, let alone verifying, properties of systems involving multiple programming languages is inherently challenging. We introduce Heterogeneous Dynamic Logic (HDL),…
An Information-Flow Perspective on Algorithmic Fairness
Samuel Teuber, Bernhard Beckert
This work presents insights gained by investigating the relationship between algorithmic fairness and the concept of secure information flow. The problem of enforcing secure inform…
Of Good Demons and Bad Angels: Guaranteeing Safe Control under Finite Precision
Samuel Teuber, Debasmita Lohar, Bernhard Beckert
As neural networks (NNs) become increasingly prevalent in safety-critical neural network-controlled cyber-physical systems (NNCSs), formally guaranteeing their safety becomes cruci…