Evaluation of Domain Agnostic Approaches for Enumeration of Minimal Unsatisfiable Subsets

Bendík, Jaroslav; Černá, Ivana

doi:10.29007/sxzb

Cited by 7 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other domain-agnostic algorithms for all cores computations over monotonic criteria appear in the literature, see, e.g., [17]. In [3] there is a comparison of several such algorithms, which concludes that none of the known algorithms is better than the others in all domains. Recently, MUST [4] was proposed as an algorithm and tool that outperforms previous ones.…”

Section: B All Unrealizable Cores For Gr(1)mentioning

confidence: 98%

Unrealizable Cores for Reactive Systems Specifications

Maoz

Shalom

2021

2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE)

View full text Add to dashboard Cite

One of the main challenges of reactive synthesis, an automated procedure to obtain a correct-by-construction reactive system, is to deal with unrealizable specifications. One means to deal with unrealizability, in the context of GR(1), an expressive assume-guarantee fragment of LTL that enables efficient synthesis, is the computation of an unrealizable core, which can be viewed as a fault-localization approach. Existing solutions, however, are computationally costly, are limited to computing a single core, and do not correctly support specifications with constructs beyond pure GR(1) elements.In this work we address these limitations. First, we present QuickCore, a novel algorithm that accelerates unrealizable core computations by relying on the monotonicity of unrealizability, on an incremental computation, and on additional properties of GR(1) specifications. Second, we present Punch, a novel algorithm to efficiently compute all unrealizable cores of a specification. Finally, we present means to correctly handle specifications that include higher-level constructs beyond pure GR(1) elements.We implemented our ideas on top of Spectra, an opensource language and synthesis environment. Our evaluation over benchmarks from the literature shows that QuickCore is in most cases faster than previous algorithms, and that its relative advantage grows with scale. Moreover, we found that most specifications include more than one core, and that Punch finds all the cores significantly faster than a competing naive algorithm.

show abstract

Section: B All Unrealizable Cores For Gr(1)mentioning

confidence: 98%

Unrealizable Cores for Reactive Systems Specifications

Maoz

Shalom

2021

2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE)

View full text Add to dashboard Cite

show abstract

“…For a more elaborated description of the three algorithms, please refer to the original papers [22,7,9] or to our recent work [8] where we have experimentally compared the algorithms in various constraint domains.…”

Section: Seed-shrink Schemementioning

confidence: 99%

“…Also this collection has been already used in several works, e.g. in the work by Cimatti et al [13] or in our recent papers [9,8]. The benchmarks range in their size from 70 to 16 million constraints and use from 26 to 4.4 million variables.…”

Section: Benchmarksmentioning

confidence: 99%

“…Yet, authors of Looney indeed acknowledge in their paper that Looney can handle only small input constraint sets containing just low tens of constraints. On the other hand, MUST was shown [8] to be able to efficiently work with hundreds of constraints.…”

Section: Case Studymentioning

confidence: 99%

“…Their tool currently supports the SAT and the SMT domains and can be relatively easily extended to support also another constraint domains. However, our recent evaluation [8] of contemporary domain agnostic algorithms in various constraint domains has shown that the efficiency of the algorithms (including MARCO) varies a lot in different constraint domains. There is no silver bullet algorithm that would be efficient in all the domains.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MUST: Minimal Unsatisfiable Subsets Enumeration Tool

Bendík

Černá

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

In many areas of computer science, we are given an unsatisfiable set of constraints with the goal to provide an insight into the unsatisfiability. One of common approaches is to identify minimal unsatisfiable subsets (MUSes) of the constraint set. The more MUSes are identified, the better insight is obtained. However, since there can be up to exponentially many MUSes, their complete enumeration might be intractable. Therefore, we focus on algorithms that enumerate MUSes online, i.e. one by one, and thus can find at least some MUSes even in the intractable cases. Since MUSes find applications in different constraint domains and new applications still arise, there have been proposed several domain agnostic algorithms. Such algorithms can be applied in any constraint domain and thus theoretically serve as ready-to-use solutions for all the emerging applications. However, there are almost no domain agnostic tools, i.e. tools that both implement domain agnostic algorithms and can be easily extended to support any constraint domain. In this work, we close this gap by introducing a domain agnostic tool called MUST. Our tool outperforms other existing domain agnostic tools and moreover, it is even competitive to fully domain specific solutions.

show abstract

Hashing-based approximate counting of minimal unsatisfiable subsets

Bendík

Meel

2023

Form Methods Syst Des

View full text Add to dashboard Cite

Evaluation of Domain Agnostic Approaches for Enumeration of Minimal Unsatisfiable Subsets

Cited by 7 publications

References 25 publications

Unrealizable Cores for Reactive Systems Specifications

Unrealizable Cores for Reactive Systems Specifications

MUST: Minimal Unsatisfiable Subsets Enumeration Tool

Hashing-based approximate counting of minimal unsatisfiable subsets

Contact Info

Product

Resources

About