Verifying Neural Network Controlled Systems Using Neural Networks

DOI10.1145/3501710.3519511MaRDI QIDQ6120690zbMATH OpenOpenAlexWikidataFDO

Authors

Publication date 21 February 2024

Published in 25th ACM International Conference on Hybrid Systems: Computation and Control (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1145/3501710.3519511

zbMATH Keywords

barrier certificatesmixed integer programmingneural networksneural-network-controlled systemssafety verification

Mathematics Subject Classification ID

Specification and verification (program logics, model checking, etc.) (68Q60) Learning and adaptive systems in artificial intelligence (68T05) Control/observation systems governed by functional relations other than differential equations (such as hybrid and switching systems) (93C30)

Recommendations

• Learning safe neural network controllers with barrier certificates
• Learning safe neural network controllers with barrier certificates
• Formal verification of neural network controlled autonomous systems
• Verisig
• Automated and formal synthesis of neural barrier certificates for dynamical models

Cites work

• A Framework for Worst-Case and Stochastic Safety Verification Using Barrier Certificates
• A method for invariant generation for polynomial continuous systems
• Approximation by superpositions of a sigmoidal function
• Automated and formal synthesis of neural barrier certificates for dynamical models
• Automatic invariant generation for hybrid systems using ideal fixed points
• Case study
• Constructing invariants for hybrid systems
• Deep learning
• dReal: an SMT solver for nonlinear theories over the reals
• Exponential-condition-based barrier certificate generation for safety verification of hybrid systems
• Formal verification of neural network controlled autonomous systems
• scientific article; zbMATH DE number 2107836 (Why is no real title available?)
• scientific article; zbMATH DE number 869736 (Why is no real title available?)
• Hybrid Systems: Computation and Control
• Learning safe neural network controllers with barrier certificates
• Model predictive control: Theory and practice - a survey
• Multilayer feedforward networks are universal approximators
• Nonparametric guidance of autoencoder representations using label information
• Reluplex: an efficient SMT solver for verifying deep neural networks
• Synthesizing barrier certificates using neural networks

Cited in

(5)

• Formal verification of neural network controlled autonomous systems
• Verisig
• Gray-box adversarial testing for control systems with machine learning components
• A learner-verifier framework for neural network controllers and certificates of stochastic systems
• Iteratively synthesizing \(\epsilon\)-robust barrier certificates for neural network controlled systems

This page was built for publication: Verifying Neural Network Controlled Systems Using Neural Networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6120690)