Exploiting policy knowledge in online least-squares policy iteration: An empirical study

peer reviewed