Comparing learning algorithms in automated assume-guarantee reasoning
Journal
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Journal Volume
6415 LNCS
Journal Issue
PART 1
Pages
643-657
Date Issued
2010
Author(s)
Abstract
We compare two learning algorithms for generating contextual assumptions in automated assume-guarantee reasoning. The CDNF algorithm implicitly represents contextual assumptions by a conjunction of DNF formulae, while the OBDD learning algorithm uses ordered binary decision diagrams as its representation. Using these learning algorithms, the performance of assume-guarantee reasoning is compared with monolithic interpolation-based Model Checking in parametrized hardware test cases. © 2010 Springer-Verlag.
Type
book