We provide a wide-ranging study of the scenario where a subset of the relations in a relational vocabulary is visible to a user—that is, their complete contents are known—while the remaining relations are invisible. We also have a background theory—invariants given by logical sentences—that may relate the visible relations to invisible ones, and also may constrain both the visible and invisible relations in isolation. We want to determine whether some other information, given as a positive existential formula, can be inferred using only the visible information and the background theory. This formula whose inference we are concerned with is denoted as the query . We consider whether positive information about the query can be inferred, and also whether negative information—the sentence does not hold—can be inferred. We further consider both the instance-level version of the problem, where both the query and the visible instance are given, and the schema-level version, where we want to know whether truth or falsity of the query can be inferred in some instance of the schema.

We look at characterizing which formulas are expressible in rich decidable logics such as guarded fixpoint logic, unary negation fixpoint logic, and guarded negation fixpoint logic. We consider semantic characterizations of definability, as well as effective characterizations. Our algorithms revolve around a finer analysis of the tree-model property and a refinement of the method of moving back and forth between relational logics and logics over trees.

We consider entailment problems involving powerful constraint languages such as frontier-guarded existential rules in which we impose additional semantic restrictions on a set of distinguished relations. We consider restricting a relation to be transitive, restricting a relation to be the transitive closure of another relation, and restricting a relation to be a linear order. We give some natural variants of guardedness that allow inference to be decidable in each case, and isolate the complexity of the corresponding decision problems. Finally we show that slight changes in these conditions lead to undecidability.

We look at characterizing which formulas are expressible in rich decidable logics such as guarded fixpoint logic, unary negation fixpoint logic, and guarded negation fixpoint logic. We consider semantic characterizations of definability, as well as effective characterizations. Our algorithms revolve around a finer analysis of the tree-model property and a refinement of the method of moving back-and-forth between relational logics and logics over trees.

Guardedness restrictions are one of the principal means to obtain decidable logics --- operators such as negation are restricted so that the free variables are contained in an atom. While guardedness has been applied fruitfully in the setting of first-order logic, the ability to add fixpoints while retaining decidability has been very limited. Here we show that one of the main restrictions imposed in the past can be lifted, getting a richer decidable logic by allowing fixpoints in which the parameters of the fixpoint can be unguarded. Using automata, we show that the resulting logics have a decidable satisfiability problem, and provide a fine study of the complexity of satisfiability. We show that similar methods apply to decide questions concerning the elimination of fixpoints within formulas of the logic.

Desirable properties of a logic include decidability, and a model theory that inherits properties of first-order logic, such as interpolation and preservation theorems. It is known that the Guarded Fragment (GF) of first-order logic is decidable and satisfies some preservation properties from first-order model theory; however, it fails to have Craig interpolation. The Guarded Negation Fragment (GNF), a recently defined extension, is known to be decidable and to have Craig interpolation. Here we give the first results on effective interpolation for extensions of GF. We provide an interpolation procedure for GNF whose complexity matches the doubly exponential upper bound for satisfiability of GNF. We show that the same construction gives not only Craig interpolation, but Lyndon interpolation and relativized interpolation, which can be used to provide effective proofs of some preservation theorems. We provide upper bounds on the size of GNF interpolants for both GNF and GF input, and complement this with matching lower bounds.

A logic satisfies Craig interpolation if whenever one formula ?1 in the logic entails another formula ?2 in the logic, there is an intermediate formula -- one entailed by ?1 and entailing ?2 -- using only relations in the common signature of ? and ?2. Uniform interpolation strengthens this by requiring the interpolant to depend only on ?1 and the common signature. A uniform interpolant can thus be thought of as a minimal upper approximation of a formula within a sub signature. For first-order logic, interpolation holds but uniform interpolation fails. Uniform interpolation is known to hold for several modal and description logics, but little is known about uniform interpolation for fragments of predicate logic over relations with arbitrary arity. Further, little is known about ordinary Craig interpolation for logics over relations of arbitrary arity that have a recursion mechanism, such as fix point logics. In this work we take a step towards filling these gaps, proving interpolation for a decidable fragment of least fix point logic called unary negation fix point logic. We prove this by showing that for any fixed k, uniform interpolation holds for the k-variable fragment of the logic. In order to show this we develop the technique of reducing questions about logics with tree-like models to questions about modal logics, following an approach by Gradel, Hirsch, and Otto. While this technique has been applied to expressivity and satisfiability questions before, we show how to extend it to reduce interpolation questions about such logics to interpolation for the µ-calculus.

Given a formula phi(x, X) positive in X, the bounded ness problem asks whether the fix point induced by phi is reached within some uniform bound independent of the structure (i.e. Whether the fix point is spurious, and can in fact be captured by a finite unfolding of the formula). In this paper, we study the bounded ness problem when phi is in the guarded fragment or guarded negation fragment of first-order logic, or the fix point extensions of these logics. It is known that guarded logics have many desirable computational and model theoretic properties, including in some cases decidable bounded ness. We prove that bounded ness for the guarded negation fragment is decidable in elementary time, and, making use of an unpublished result of Colcombet, even 2EXPTIME-complete. Our proof extends the connection between guarded logics and automata, reducing bounded ness for guarded logics to a question about cost automata on trees, a type of automaton with counters that assigns a natural number to each input rather than just a boolean.

Weakly definable languages of infinite trees are an expressive subclass of regular tree languages definable in terms of weak monadic second-order logic, or equivalently weak alternating automata. Our main result is that given a Büchi automaton, it is decidable whether the language is weakly definable. We also show that given a parity automaton, it is decidable whether the language is recognizable by a nondeterministic co-Büchi automaton. The decidability proofs build on recent results about cost automata over infinite trees. These automata use counters to define functions from infinite trees to the natural numbers extended with infinity. We reduce to testing whether the functions defined by certain "quasi-weak" cost automata are bounded by a finite value.

Cost automata have a finite set of counters which can be manipulated on each transition but do not affect control flow. Based on the evolution of the counter values, these automata define functions from a domain like words or trees to \N \cup \set{\infty}, modulo an equivalence relation which ignores exact values but preserves boundedness properties. These automata have been studied by Colcombet et al. as part of a "theory of regular cost functions", an extension of the theory of regular languages which retains robust equivalences, closure properties, and decidability like the classical theory. We extend this theory by introducing quasi-weak cost automata. Unlike traditional weak automata which have a hard-coded bound on the number of alternations between accepting and rejecting states, quasi-weak automata bound the alternations using the counter values (which can vary across runs). We show that these automata are strictly more expressive than weak cost automata over infinite trees. The main result is a Rabin-style characterization theorem: a function is quasi-weak definable if and only if it is definable using two dual forms of non-deterministic Büchi cost automata. This yields a new decidability result for cost functions over infinite trees.

Abstract In [3]. two different effective versions of Borel embedding are defined. The first, called computable embedding, is based on uniform enumeration reducibility. while the second, called Turing computable embedding, is based on uniform Turing reducibility. While [3] focused mainly on computable embeddings, the present paper considers Turing computable embeddings. Although the two notions are not equivalent, we can show that they behave alike on the mathematically interesting classes chosen for investigation in [3]. We give a “Pull-back Theorem”, saying that if Ф is a Turing computable embedding of K into K′, then for any computable infinitary sentence φ in the language of K′, we can find a computable infinitary sentence φ* in the language of K such that for all A ∈ K A ⊨ φ* iff Φ (A) ⊨ φ and φ* has the same “complexity” as φ (i.e., if φ is computable Σα or computable Πα, for α ≥ 1, then so is φ*). The Pull-back Theorem is useful in proving non-embeddability, and it has other applications as well.