Automatic Abstraction Refinement in Neural Network Verification using Sensitivity Analysis

DOI10.1145/3575870.3587129OpenAlexW4375851954MaRDI QIDQ6202102

Matthias Althoff, Unnamed Author

Publication date: 21 February 2024

Published in: Proceedings of the 26th ACM International Conference on Hybrid Systems: Computation and Control (Search for Journal in Brave)

Full work available at URL: https://mediatum.ub.tum.de/doc/1710698/document.pdf

zbMATH Keywords

sensitivity analysisneural networksformal verificationset-based computingautomatic abstraction refinementpolynomial zonotopes.

Mathematics Subject Classification ID

Formal languages and automata (68Q45) Specification and verification (program logics, model checking, etc.) (68Q60) Control/observation systems governed by functional relations other than differential equations (such as hybrid and switching systems) (93C30)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Verisig 2.0: verification of neural network controllers using Taylor model preconditioning
• Multilayer feedforward networks are universal approximators
• Verification of piecewise deep neural networks: a star set approach with zonotope pre-filter
• Reachable sets of classifiers and regression models: (non-)robustness analysis and robust training
• Safety verification of deep neural networks
• Reluplex: an efficient SMT solver for verifying deep neural networks
• Sparse Polynomial Zonotopes: A Novel Set Representation for Reachability Analysis
• Utilizing dependencies to obtain subsets of reachable sets
• JuliaReach
• Learning representations by back-propagating errors
• Hybrid Systems: Computation and Control

This page was built for publication: Automatic Abstraction Refinement in Neural Network Verification using Sensitivity Analysis