Synthesizing ReLU neural networks with two hidden layers as barrier certificates for hybrid systems

DOI10.1145/3447928.3456638OpenAlexW3157818965MaRDI QIDQ6120657

No author found.

Publication date: 21 February 2024

Published in: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1145/3447928.3456638

zbMATH Keywords

hybrid systemsneural networksmixed integer programmingsafety verificationbarrier certificates

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
• Barrier certificates revisited
• Computing differential invariants of hybrid systems as fixed points
• Multilayer feedforward networks are universal approximators
• Automated and formal synthesis of neural barrier certificates for dynamical models
• Reluplex: an efficient SMT solver for verifying deep neural networks
• An efficient framework for barrier certificate generation of uncertain nonlinear hybrid systems
• Reachability analysis using message passing over tree decompositions
• Constructing invariants for hybrid systems
• A Method for Invariant Generation for Polynomial Continuous Systems
• Exponential-Condition-Based Barrier Certificate Generation for Safety Verification of Hybrid Systems
• Automatic invariant generation for hybrid systems using ideal fixed points
• Compositional safety analysis using barrier certificates
• Simulation-guided lyapunov analysis for hybrid dynamical systems
• Discovering Forward Invariant Sets for Nonlinear Dynamical Systems
• Bounded Verification with On-the-Fly Discrepancy Computation
• Constraint-Based Approach for Analysis of Hybrid Systems
• Generating Invariants for Non-linear Hybrid Systems by Linear Algebraic Methods
• Synthesizing barrier certificates using neural networks
• Sherlock - A tool for verification of neural network feedback systems
• Verification and synthesis using real quantifier elimination
• A Framework for Worst-Case and Stochastic Safety Verification Using Barrier Certificates
• Hybrid Systems: Computation and Control
• Hybrid Systems: Computation and Control
• Approximation by superpositions of a sigmoidal function

This page was built for publication: Synthesizing ReLU neural networks with two hidden layers as barrier certificates for hybrid systems