Study of sample efficiency improvements for reinforcement learning algorithms | IEEE Conference Publication | IEEE Xplore