Automata-learning includes techniques to construct (learn/identify) model of a given -black box- system in which its underlying model is an automaton in an automatic way.
Like other -machine- learning approaches, it can be categorized as offline/passive and online/active way. The main focus of AIDE is to provide active automata-learning in which there is a teacher (also called oracle) who knows the model of system and is able to answer two kinds of queries: membership queries and equivalence queries.
Note that in black-box system, we do not have the underlying model of system and so, the there is no real oracle who can answer equivalence queries. So, we should provide a way to approximate the equivalence queries. The solution is usually using testing to approximate the equivalence queries.
- Changes to previous version:
Initial Announcement on mloss.org.
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.