A note on the query complexity of learning DFA

Abstract

It is known that the class of deterministic finite automata is polynomial time learnable by using membership and equivalence queries. We investigate -- the query complexity -- the number of membership and equivalence queries for learning deterministic finite automata. First we consider two different learning situations and show lower bounds on the query complexity of learning deterministic finite automata. We then investigate the query complexity in general setting, and show some trade-off phenomenon between the number of membership and equivalence queries.