A Formally Verified IEEE 754 Floating-Point Implementation of Interval Iteration for MDPs

Bram Kohlen, Maximilian Schäffeler, Mohammad Abdulaziz, Arnd Hartmanns, Peter Lammich · 2025

Read A Formally Verified IEEE 754 Floating-Point Implementation of Interval Iteration for MDPs by Bram Kohlen, Maximilian Schäffeler, Mohammad Abdulaziz, Arnd Hartmanns, Peter Lammich (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-10127.txt.

Description

Bram Kohlen, Maximilian Schäffeler, Mohammad Abdulaziz, Arnd Hartmanns, and Peter Lammich wrote a 2025 computer science paper on a formally verified tool for analyzing Markov decision processes. These models describe systems that act under uncertainty, and interval iteration is a standard way to compute their long run values. Floating point rounding, though, can quietly corrupt such results.

The team built an implementation whose correctness is proven end to end, from abstract math down to machine code with real floating point arithmetic. They used the Isabelle proof assistant and extended its refinement framework to handle rounding modes. The verified tool ran at speeds competitive with leading unverified tools. Users get both performance and a formal guarantee, a rare pairing in numerical software.

Catalog record

CREATORBram Kohlen, Maximilian Schäffeler, Mohammad Abdulaziz, Arnd Hartmanns, Peter Lammich
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE71 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-10127

Browse the full library · About the Tower of Babel