Markov decision processes (MDP) are useful to model concurrent process optimisation problems, but verifying them with numerical methods is often intractable. Existing approximative approaches do not scale well and are limited to memoryless schedulers. Here we present the basis of scalable verification for MDPSs, using an O(1) memory representation of history-dependent schedulers. We thus facilitate scalable learning techniques and the use of massively parallel verification.
Abstract. Last time we reported on Romeo, analyses with this tool were mostly based on translations to other tools. This new version provides an integrated TCTL model-checker and has gained in expressivity with the addition of parameters. Although there exists other tools to compute the state-space of stopwatch models, Romeo is the first one that performs TCTL model-checking on stopwatch models. Moreover, it is the first tool that performs TCTL model-checking on timed parametric models. Indeed, Romeo now features an efficient model-checking of time Petri nets using the Uppaal DBM Library, the model-checking of stopwatch Petri nets and parametric stopwatch Petri nets using the Parma Polyhedra Library and a graphical editor and simulator of these models. Furthermore, its audience has increased leading to several industrial contracts. This paper reports on these recent developments of Romeo.
Markov decision processes (MDP) are useful to model optimisation problems in concurrent systems. To verify MDPs with efficient Monte Carlo techniques requires that their nondeterminism be resolved by a scheduler. Recent work has introduced the elements of lightweight techniques to sample directly from scheduler space, but finding optimal schedulers by simple sampling may be inefficient. Here we describe "smart" sampling algorithms that can make substantial improvements in performance.
International audienceQuantitative security analysis evaluates and compares how effectively a system protects its secret data. We introduce QUAIL, the first tool able to perform an arbitrary-precision quantitative analysis of the security of a system depending on private information. QUAIL builds a Markov Chain model of the system's behavior as observed by an attacker, and computes the correlation between the system's observable output and the behavior depending on the private information, obtaining the expected amount of bits of the secret that the attacker will infer by observing the system. QUAIL is able to evaluate the safety of randomized protocols depending on secret data, allowing to verify a security protocol's effectiveness. We experiment with a few examples and show that QUAIL's security analysis is more accurate and revealing than results of other tools
Abstract. Specification theories for real-time systems allow to reason about interfaces and their implementation models, using a set of operators that includes satisfaction, refinement, logical and parallel composition. To make such theories applicable throughout the entire design process from an abstract specification to an implementation, we need to be able to reason about possibility to effectively implement the theoretical specifications on physical systems. In the literature, this implementation problem has already been linked to the robustness problem for Timed Automata, where small perturbations in the timings of the models are introduced. We address the problem of robust implementations in timed specification theories. Our contributions include the analysis of robust timed games and the study of robustness with respect to the operators of the theory.
Abstract. We present an overview of Plasma Lab, a modular statistical model checking (SMC) platform that facilitates multiple SMC algorithms, multiple modelling and query languages and has multiple modes of use. Plasma Lab may be used as a stand-alone tool with a graphical development environment or invoked from the command line for high performance scripting applications. Plasma Lab is written in Java for maximum cross-platform compatibility, but it may interface with tools and libraries written in arbitrary programming languages. Plasma Lab's API also allows it to be incorporated as a library within other tools. We first describe the motivation and architecture of Plasma Lab, then proceed to describe some of its important algorithms, including those for rare events and nondeterminism. We conclude with a number of industrially-relevant case studies and applications.
International audienceWe introduce a new notion of structural refinement, a sound abstraction of logical implication, for the modal nu-calculus. Using new translations between the modal nu-calculus and disjunctive modal transition systems, we show that these two specification formalisms are structurally equivalent.Using our translations, we also transfer the structural operations of composition and quotient from disjunctive modal transition systems to the modal nu-calculus. This shows that the modal nu-calculus supports composition and decomposition of specifications
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.