Abstract:Probabilistic model checking is a verification technique that has been the focus of intensive research for over a decade. One important issue with probabilistic model checking, which is crucial for its practical significance but is overlooked by the state-of-the-art largely, is the potential discrepancy between a stochastic model and the real-world system it represents when the model is built from statistical data. In the worst case, a tiny but nontrivial change to some model quantities might lead to misleading or even invalid verification results. To address this issue, in this paper, we present a mathematical characterization of the consequences of model perturbations on the verification distance. The formal model that we adopt is a parametric variant of discrete-time Markov chains equipped with a vector norm to measure the perturbation. Our main technical contributions include a closed-form formulation of asymptotic perturbation bounds, and computational methods for two arguably most useful forms of those bounds, namely linear bounds and quadratic bounds. We focus on verification of reachability properties but also address automata-based verification of omega-regular properties. We present the results of a selection of case studies that demonstrate that asymptotic perturbation bounds can accurately estimate maximum variations of verification results induced by model perturbations.

Perturbation Analysis in Verification of Discrete-Time Markov Chains.

Asymptotic Perturbation Bounds for Probabilistic Model Checking with Empirically Determined Probability Parameters

The Arsenal of Perturbation Bounds for Finite Continuous-Time Markov Chains: A Perspective

Perturbations of Markov Chains

Blockwise Perturbation Theory for Markov Chains

Perturbation Analysis for Finite-Time Stability and Stabilization of Probabilistic Boolean Networks.

Stochastic Stability of Genetic Networks with Disturbance Attenuation.

Verification of Linear Duration Properties over Continuous-Time Markov Chains

Perturbation of error bounds

Probabilistic perturbation bounds of matrix decompositions

Stability and Stabilization of Uncertain 2-D Discrete Systems with Stochastic Perturbation

Perturbations in Networked Control Systems

Relationship Between Perturbation Realization Factors with Queueing Models and Markov Models

Perturbation and Inverse Problems of Stochastic Matrices

Perturbation theory for killed Markov processes and quasi-stationary distributions

Blockwise perturbation theory for nearly uncoupled Markov chains and its application

A stochastic perturbation analysis of the QR decomposition and its applications

Finite Verification for Robustness of Discrete Systems under Structured Perturbations

Model Checking Quantum Markov Chains

Metastability of Exponentially Perturbed Markov Chains

Model Checking of Continuous-Time Markov Chains Against Timed Automata Specifications