Automatic data refinement

  • Authors:
  • Peter Lammich

  • Affiliations:
  • Technische Universität München, Germany

  • Venue:
  • ITP'13 Proceedings of the 4th international conference on Interactive Theorem Proving
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present the Autoref tool for Isabelle/HOL, which automatically refines algorithms specified over abstract concepts like maps and sets to algorithms over concrete implementations like red-black-trees, and produces a refinement theorem. It is based on ideas borrowed from relational parametricity due to Reynolds and Wadler. The tool allows for rapid prototyping of verified, executable algorithms. Moreover, it can be configured to fine-tune the result to the user's needs. Our tool is able to automatically instantiate generic algorithms, which greatly simplifies the implementation of executable data structures. Thanks to its integration with the Isabelle Refinement Framework and the Isabelle Collection Framework, Autoref can be used as a backend to a stepwise refinement based development approach, having access to a rich library of verified data structures. We have evaluated the tool by synthesizing efficiently executable refinements for some complex algorithms, as well as by implementing a library of generic algorithms for maps and sets.