Article ID Journal Published Year Pages File Type
423569 Electronic Notes in Theoretical Computer Science 2016 16 Pages PDF
Abstract

Reversible computing is bi-deterministic which means that its execution is both forward and backward deterministic, i.e. next/previous computational step is uniquely determined. Various approaches exist to catch its extensional or intensional aspects and properties. We present a class RPRF of reversible functions which holds at bay intensional aspects and emphasizes the extensional side of the reversible computation by following the style of Dedekind-Robinson Primitive Recursive Functions. The class RPRF is closed by inversion, can only express bijections on integers — not only natural numbers —, and it is expressive enough to simulate Primitive Recursive Functions, of course, in an effective way.

Related Topics
Physical Sciences and Engineering Computer Science Computational Theory and Mathematics