Project description
Software testing based on finite-state systems
Determining what can programmes find out algorithmically about other programmes remains a fundamental question in theoretical computer science. By restricting the class of analysed programmes, recent studies have demonstrated a wide variety of methods for checking various important properties. Understanding how to restrict the analysed programmes is crucial in software verification as algorithms can be used to automatically check software correctness. However, designing models that can work with infinite-state systems to account for how the programme behaves for all possible inputs poses a challenge. The ERC-funded FINABIS project will investigate how to transform infinite-state systems into finite-state systems that preserve some pertinent aspects of the original system. Researchers will study two specific transformations: separability and closure problems.
Objective
The algorithmic analysis of infinite-state systems is a central topic
of theoretical computer science that is part of a popular approach to
software verification. While analyzing infinite-state systems is
indispensable when verifying software, finite-state sytems are far
better understood and permit much more efficient analysis. In this
project, I will pursue fundamental questions that arise when we want
to abstract infinite-state systems by finite-state systems. The goal
is to understand two types of problems:
1. Separability problems: Given two infinite-state systems, can we
find a finite-state overapproximation of the first system whose
behaviors are disjoint from those of the second system? Separability
is a basic task for synthesizing certificates for disjointness and
therefore safety properties in concurrent systems.
2. Closure computation. There are several non-constructive results
that guarantee the existence of finite-state overapproximations of
infinite-state systems that preserve some particular information. We
are interested in how to compute these overapproximations effectively
and efficiently. Examples include downward closures and upward
closures with respect to the (scattered) subword ordering. Efficient
procedures for closure computation would have immediate implications
for infinite-state verification tasks that combine recursion with
concurrency.
In addition to directly attacking well-known deep open problems
regarding these fundamental questions, the project will also develop
methods that will likely be crucial for resolving further major open
problems in infinite-state systems. Moreover, the obtained results
would have immediate implications for software verification in
settings that combine recursion with concurrency, which is a
notoriously difficult task.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
You need to log in or register to use this function
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Topic(s)
Funding Scheme
HORIZON-ERC - HORIZON ERC GrantsHost institution
80539 Munchen
Germany