Rigid E-unification
نویسنده
چکیده
By replacing syntactical unification with rigid E-unification, equality handling can be added to rigid variable calculi for first-order logic, including free variable tableau (Fitting, 1996), the mating method (Andrews, 1981), the connection method (Bibel, 1982), and model elimination (Loveland, 1969); for an overview of these calculi, see Chapters I.1.1 and I.1.2. Rigid E-unification and its significance for automated theorem proving was first described in (Gallier et al., 1987). An earlier attempt to formulate the generalized unification problem that has to be solved for handling equality in rigid variable calculi can be found in (Bibel, 1982). Ground E-unification (i.e., E-unification with variable-free equalities) has long been known to be decidable (Sect. 2.3), and classical universal E-unification has long been known to be undecidable (Chap. I.2.7). Rigid E-unification is in between: It is decidable in the simple, non-simultaneous case (Sect. 2.4), but it is undecidable whether there is a simultaneous solution for several rigid E-unification problems (Sect. 3.2), which is unfortunate as simultaneous rigid E-unification is of great importance for handling equality in automated theorem proving (Sect. 5). In the remainder of this section, we describe the basic idea of rigid Eunification and its importance for adding equality to rigid variable calculi and introduce syntax and semantics of first-order logic with equality. In Section 2, we formally define (non-simultaneous) rigid E-unification and the notion of (minimal) complete sets of unifiers; and we briefly sketch proofs for the decidability of ground E-unification and—based on this—for rigid E-unification; methods for solving rigid E-unification problems are compared. In Section 3.3, the problem of finding a simultaneous solution for several rigid E-unification problems is discussed; and in Section 4, mixed E-unification is introduced, that is a combination of classical and rigid E-unification. Using the example of free variable semantic tableaux, we show in Section 5 how rigid E-unification can be used to handle equality in a rigid variable calcu-
منابع مشابه
Bernhard Beckert Rigid E - Unification
By replacing syntactical unification with rigid E-unification, equality handling can be added to rigid variable calculi for first-order logic, including free variable tableau (Fitting, 1996), the mating method (Andrews, 1981), the connection method (Bibel, 1982), and model elimination (Loveland, 1969); for an overview of these calculi, see Chapters I.1.1 and I.1.2. Rigid E-unification and its s...
متن کاملA Completion-Based Method for Mixed Universal and Rigid E-Unification
We present a completion-based method for handling a new version of E-unification, called “mixed” E-unification, that is a combination of the classical “universal” E-unification and “rigid” E-unification. Rigid E-unification is an important method for handling equality in Gentzen-type first-order calculi, such as free-variable semantic tableaux or matings. The performance of provers using E-unif...
متن کاملImplementing rigid e-unification
Rigid E-unification problems arise naturally in automated theorem provers that deal with equality. While there is a lot of theory about rigid E-unification, only few implementations exist. Since the problem is NPcomplete, direct implementations of the theory are slow. In this paper we discuss how to implement a rigid E-unifier, focussing on efficiency. First, we introduce an efficient represent...
متن کاملEfficient Algorithms for Bounded Rigid E-unification
Rigid E -unification is the problem of unifying two expressions modulo a set of equations, with the assumption that every variable denotes exactly one term (rigid semantics). This form of unification was originally developed as an approach to integrate equational reasoning in tableau-like proof procedures, and studied extensively in the late 80s and 90s. However, the fact that simultaneous rigi...
متن کاملFree Variables and Theories: Revisiting Rigid E-unification
The efficient integration of theory reasoning in first-order calculi with free variables (such as sequent calculi or tableaux) is a longstanding challenge. For the case of the theory of equality, an approach that has been extensively studied in the 90s is rigid E -unification, a variant of equational unification in which the assumption is made that every variable denotes exactly one term (rigid...
متن کامل