Learning Buechi Automata and Its Applications

摘要

In this work, we review an algorithm that learns a Biichi automaton from a teacher who knows an ω-regular language; the algorithm is based on learning a formalism named family of DFAs (FDFAs) recently proposed by Angluin and Fisman. We introduce the learning algorithm by learning the simple ω-regular language (ab)~ω: besides giving the readers an overview of the algorithm, it guides them on how the algorithm works step by step. Further, we demonstrate how the learning algorithm can be exploited in classical automata operations such as complementation checking and in the context of termination analysis.