AIGEN: Random Generation of Symbolic Transition Systems
Authors:
Open Access
- 15 July 2021
- book chapter
- Published by Springer Nature
- p. 435-446
Abstract
AIGEN is an open source tool for the generation of transition systems in a symbolic representation. To ensure diversity, it employs a uniform random sampling over the space of all Boolean functions with a given number of variables. AIGEN relies on reduced ordered binary decision diagrams (ROBDDs) and canonical disjunctive normal form (CDNF) as canonical representations that allow us to enumerate Boolean functions, in the former case with an encoding that is inspired by data structures used to implement ROBDDs. Several parameters allow the user to restrict generation to Boolean functions or transition systems with certain properties, which are then output in AIGER format. We report on the use of AIGEN to generate random benchmark problems for the reactive synthesis competition SYNTCOMP 2019, and present a comparison of the two encodings with respect to time and memory efficiency in practice.Keywords
This publication has 6 references indexed in Scilit:
- AIGEN: Random Generation of Symbolic Boolean Functions and Transition Systems2021
- A Symbolic Algorithm for Lazy Synthesis of Eager Strategies2018
- The first reactive synthesis competition (SYNTCOMP 2014)International Journal on Software Tools for Technology Transfer, 2016
- Competition on Software Verification2012
- ABC: An Academic Industrial-Strength Verification Tool2010
- Design and Results of the First Satisfiability Modulo Theories Competition (SMT-COMP 2005)Journal of Automated Reasoning, 2005