Finite learning by a “team”