Login / Signup

The probabilistic termination tool amber.

Marcel MoosbruggerEzio BartocciJoost-Pieter KatoenLaura Kovács
Published in: Formal methods in system design (2023)
We describe the Amber tool for proving and refuting the termination of a class of probabilistic while-programs with polynomial arithmetic, in a fully automated manner. Amber combines martingale theory with properties of asymptotic bounding functions and implements relaxed versions of existing probabilistic termination proof rules to prove/disprove (positive) almost sure termination of probabilistic loops. Amber supports programs parametrized by symbolic constants and drawing from common probability distributions. Our experimental comparisons give practical evidence of Amber outperforming existing state-of-the-art tools.
Keyphrases
  • public health
  • machine learning
  • single cell