Risk verification of stochastic systems with neural network controllers

DOI10.1016/J.ARTINT.2022.103782MaRDI QIDQ2093383zbMATH OpenOpenAlexFDO

Authors Matthew Cleaveland, Lars Lindemann, Radoslav Ivanov, George Pappas

Publication date 8 November 2022

Published in Artificial Intelligence (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2209.09881

zbMATH Keywords

stochastic systemsformal verificationneural network controller verificationrisk-aware systems

Mathematics Subject Classification ID

Artificial intelligence (68Txx)

Recommendations

• A learner-verifier framework for neural network controllers and certificates of stochastic systems
• Generating probabilistic safety guarantees for neural network controllers
• Verisig
• Stochastic sampled-data stabilization of neural-network-based control systems
• Formal verification of neural network controlled autonomous systems
• Stable neural control of uncertain multivariable systems
• RINO: Robust INner and Outer Approximated Reachability of Neural Networks Controlled Systems
• Neural network-based asymptotic tracking control design for stochastic nonlinear systems
• Adaptive neural network control of stochastic nonlinear systems with perturbations

Cites work

• scientific article; zbMATH DE number 1461253 (Why is no real title available?)
• scientific article; zbMATH DE number 5585443 (Why is no real title available?)
• Being Correct Is Not Enough: Efficient Verification Using Robust Linear Temporal Logic
• Calculation of the Wasserstein Distance Between Probability Distributions on the Line
• Case study
• Data-driven verification of stochastic linear systems with signal temporal logic constraints
• Distribution-free, Risk-controlling Prediction Sets
• Formal Techniques, Modelling and Analysis of Timed and Fault-Tolerant Systems
• GAZEBO classic
• Infinite dimensional analysis. A hitchhiker's guide.
• Reluplex: an efficient SMT solver for verifying deep neural networks
• Risk-averse autonomous systems: a brief history and recent developments from the perspective of optimal control
• Risk-constrained reinforcement learning with percentile risk criteria
• Robustness of temporal logic specifications for continuous-time signals
• Sequential Tests of Statistical Hypotheses
• Sherlock - A tool for verification of neural network feedback systems
• Statistical model checking
• Statistical verification of dynamical systems using set oriented methods
• Statistical verification of learning-based cyber-physical systems
• System design of stochastic models using robustness of temporal properties
• Verisig

Describes a project that uses

Uses Software

• ReachNN
• Sherlock
• Wasserstein GAN
• GAZEBO classic

This page was built for publication: Risk verification of stochastic systems with neural network controllers

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2093383)