Recursive First-order Syntactic Unification Modulo Variable Classes
share
We present a generalization of first-order syntactic unification to a term algebra where variable indexing is part of the object language. Unlike first-order syntactic unification, the number of variables within a given problem is not finitely bound as terms can have self-symmetric subterms (modulo an index shift) allowing the construction of infinitely deep terms containing infinitely many variables, what we refer to as arithmetic progressive terms. Such constructions are related to inductive reasoning. We show that unifiability is decidable for so-called simple linear 1-loops and conjecture decidability for less restricted classes of loops.
READ FULL TEXT