- English (2) (remove)
- Encoding induction in correctness proofs of program transformations as a termination problem (2012)
- The diagram-based method to prove correctness of program transformations consists of computing complete set of (forking and commuting) diagrams, acting on sequences of standard reductions and program transformations. In many cases, the only missing step for proving correctness of a program transformation is to show the termination of the rearrangement of the sequences. Therefore we encode complete sets of diagrams as term rewriting systems and use an automated tool to show termination, which provides a further step in the automation of the inductive step in correctness proofs.
- Computing overlappings by unification in the deterministic lambda calculus LR with letrec, case, constructors, seq and variable chains (2011)
- Correctness of program transformations in extended lambda calculi with a contextual semantics is usually based on reasoning about the operational semantics which is a rewrite semantics. A successful approach to proving correctness is the combination of a context lemma with the computation of overlaps between program transformations and the reduction rules.The method is similar to the computation of critical pairs for the completion of term rewriting systems. We describe an effective unification algorithm to determine all overlaps of transformations with reduction rules for the lambda calculus LR which comprises a recursive let-expressions, constructor applications, case expressions and a seq construct for strict evaluation. The unification algorithm employs many-sorted terms, the equational theory of left-commutativity modeling multi-sets, context variables of different kinds and a mechanism for compactly representing binding chains in recursive let-expressions. As a result the algorithm computes a finite set of overlappings for the reduction rules of the calculus LR that serve as a starting point to the automatization of the analysis of program transformations.