Two Novel On-policy Reinforcement Learning Algorithms based on TD(λ)-methods | IEEE Conference Publication | IEEE Xplore