A Parameterized Unfold/Fold Transformation Framework for Definite Logic Programs

Roychoudhury, Abhik; Kumar, K. Narayan; Ramakrishnan, C. R.; Ramakrishnan, I. V.

doi:10.1007/10704567_24

Cited by 12 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the given declarative semantics. The following correctness theorem extends similar results holding for logic programs [14,40,46] to the case of logic programs with equations and disequations.…”

Section: Correctness Of the Transformation Rules Wrt The Declaratisupporting

confidence: 79%

“…the declarative semantics of logic programs based on the least Herbrand model. The proof of this correctness result is similar to the proofs of the correctness results which are presented in [14,40,46].…”

Section: Concluding Remarks and Related Worksupporting

confidence: 76%

“…By applying the enhanced transformation rules according to the Determinization Strategy we will present in Section 6, In this section we present the program transformation rules which we use for program specialization. These rules extend the unfold/fold rules considered in [14,40,46] to logic programs with atoms which denote equations and disequations between terms. The transformation rules we present in this section enhance in several respects the rules PD1-PD4 for partial deduction which we have considered in Section 3.…”

Section: Some Limitations Of Partial Deductionmentioning

confidence: 99%

“…Similarly to [14,46,40], our extended set of program transformation rules preserves the least Herbrand model semantics. For the logic language with equations and negated equations considered in this paper, we adopt the usual Prolog operational semantics with the left-to-right selection rule, where equations are evaluated by using unification.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Derivation of Efficient Logic Programs by Specialization and Reduction of Nondeterminism

Pettorossi

Proietti

Renault

2005

Higher-Order Symb Comput

View full text Add to dashboard Cite

Program specialization is a program transformation methodology which improves program efficiency by exploiting the information about the input data which are available at compile time. We show that current techniques for program specialization based on partial evaluation do not perform well on nondeterministic logic programs. We then consider a set of transformation rules which extend the ones used for partial evaluation, and we propose a strategy for guiding the application of these extended rules so to derive very efficient specialized programs. The efficiency improvements which sometimes are exponential, are due to the reduction of nondeterminism and to the fact that the computations which are performed by the initial programs in different branches of the computation trees, are performed by the specialized programs within single branches. In order to reduce nondeterminism we also make use of mode information for guiding the unfolding process. To exemplify our technique, we show that we can automatically derive very efficient matching programs and parsers for regular languages. The derivations we have performed could not have been done by previously known partial evaluation techniques.

show abstract

Section: Correctness Of the Transformation Rules Wrt The Declaratisupporting

confidence: 79%

Section: Concluding Remarks and Related Worksupporting

confidence: 76%

Section: Some Limitations Of Partial Deductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Derivation of Efficient Logic Programs by Specialization and Reduction of Nondeterminism

Pettorossi

Proietti

Renault

2005

Higher-Order Symb Comput

View full text Add to dashboard Cite

show abstract

“…combined induction-based techniques for verification of infinite families of finite state systems (e.g., a n-processor token ring, for n > 2) with model checking. We construct induction proofs using a program transformation framework [RKRR99] we developed recently, and have devised a strategy to guide the search for a proof. This strategy automatically constructs nontrivial induction proofs for verifying safety and liveness properties of several families of systems [RKR + 99].…”

mentioning

confidence: 99%

Tabled logic programming for verification and program analysis

Ramakrishnan

2000

SIGSOFT Softw. Eng. Notes

Self Cite

View full text Add to dashboard Cite

Tabled resolution for logic programs [TS86, CW96] addresses the major shortcomings of Prolog-style resolution, namely, weak termination, repeated subcomputations, and weak semantics for negation. Techniques used for program analysis and verification, such as model checking and abstract interpretation, can be implemented using tabling by treating the semantic equations as logic rules. For instance, we implemented XMC [RRR+97], an efficient model checker for alternation-free modal mu-calculus, by encoding the semantic equations as a logic program, and evaluating the program using the XSB tabled logic programming system [XSB99].Using the XMC model checker as a starting point, this project aims to.-1. exploit the power of current tabled resolution systems to construct high-performance model checkers for systems specified in various process languages and properties specified using different temporal logics; 2. extend tabling methods to constraint logic programs, thereby tackling new applications, such as verification of infinite-state and real-time systems; and 3. integrate deductive strategies (traditionally used by theorem provers) into model checkers by combining tabled resolution with program transformation techniques.We elaborate our efforts in each of these directions below.We have encoded a local model checker for linear-time temporal logic (LTL) as a logic program. A thorough performance evaluation of the LTL model checker is underway. For the mu-calculus model checker, we are investigating optimizations that, based on the given property, reduce the state space of the system being verified. We are also exploring the formulation and implementation of equivalence checking techniques such as symbolic bisimulation checking.We have implemented XMC/RT, a model checker for timed modal mu-calculus, for verification of real-time systems [DRS99]. For this, we loosely coupled a constraint solver over reals with the XSB tabled logic programming system. This platform yields a model checker for a very large class of properties with performance comparable to several wellknown tools for verifying real-time systems. A tighter integration of constraints and tabling is currently underway.We have taken a significant step towards combining deductive strategies into a model checker. In particular, we have 24This work is supported in part by NSF grants CCR-9876242, EIA-9705985, and CCR-9711385. combined induction-based techniques for verification of infinite families of finite state systems (e.g., a n-processor token ring, for n > 2) with model checking. We construct induction proofs using a program transformation framework [RKRR99] we developed recently, and have devised a strategy to guide the search for a proof. This strategy automatically constructs nontrivial induction proofs for verifying safety and liveness properties of several families of systems [RKR + 99].

show abstract

A Parameterized Unfold/Fold Transformation Framework for Definite Logic Programs

Roychoudhury

Kumar

Ramakrishnan

et al. 1999

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Given a program P , an unfold/fold program transformation system derives a sequence of programs P = P0 , P1 , : : : , Pn , s u c h t h a t Pi+1 is derived from Pi by application of either an unfolding or a folding step. Existing unfold/fold transformation systems for de nite logic programs di er from one another mainly in the kind of folding transformations they permit at each step. Some allow folding using a single (possibly recursive) clause while others permit folding using multiple non-recursive clauses. However, none allow folding using multiple recursive clauses that are drawn from some previous program in the transformation sequence. In this paper we develop a parameterized framework for unfold/fold transformations by suitably abstracting and extending the proofs of existing transformation systems. Various existing unfold/fold transformation systems can be obtained by instantiating the parameters of the framework. This framework enables us to not only understand the relative strengths and limitations of these systems but also construct new transformation systems. Speci cally we present a more general transformation system that permits folding using multiple recursive clauses that can be drawn from any previous program in the transformation sequence. This new transformation system is also obtained by instantiating our parameterized framework.

show abstract

A Parameterized Unfold/Fold Transformation Framework for Definite Logic Programs

Cited by 12 publications

References 12 publications

Derivation of Efficient Logic Programs by Specialization and Reduction of Nondeterminism

Derivation of Efficient Logic Programs by Specialization and Reduction of Nondeterminism

Tabled logic programming for verification and program analysis

A Parameterized Unfold/Fold Transformation Framework for Definite Logic Programs

Contact Info

Product

Resources

About