Comparing different methods to speed up reinforcement learning in a complex domain | IEEE Conference Publication | IEEE Xplore