Better automated importance splitting for transient rare events

Statistical model checking uses simulation to overcome the state space explosion problem in formal verification. Yet its runtime explodes when faced with rare events, unless a rare event simulation method like importance splitting is used. The effectiveness of importance splitting hinges on nontrivi...

Descripción completa

Detalles Bibliográficos
Autores: Budde, Carlos Ernesto, D'argenio, Pedro Ruben, Hartmanns, Arnd
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2017
País:Argentina
Institución:Consejo Nacional de Investigaciones Científicas y Técnicas
Repositorio:CONICET Digital (CONICET)
Idioma:inglés
OAI Identifier:oai:ri.conicet.gov.ar:11336/72324
Acceso en línea:http://hdl.handle.net/11336/72324
Access Level:acceso abierto
Palabra clave:Rare Event Simulation
Importance Splitting
Transient Analysis
Expected Success
https://purl.org/becyt/ford/1.2
https://purl.org/becyt/ford/1
Descripción
Sumario:Statistical model checking uses simulation to overcome the state space explosion problem in formal verification. Yet its runtime explodes when faced with rare events, unless a rare event simulation method like importance splitting is used. The effectiveness of importance splitting hinges on nontrivial model-specific inputs: an importance function with matching splitting thresholds. This prevents its use by non-experts for general classes of models. In this paper, we propose new method combinations with the goal of fully automating the selection of all parameters for importance splitting. We focus on transient (reachability) properties, which particularly challenged previous techniques, and present an exhaustive practical evaluation of the new approaches on case studies from the literature. We find that using Restart simulations with a compositionally constructed importance function and thresholds determined via a new expected success method most reliably succeeds and performs very well. Our implementation within the Modest Toolset supports various classes of formal stochastic models and is publicly available.