Mixed Reinforcement Learning for Efficient Policy Optimization in Stochastic Environments | IEEE Conference Publication | IEEE Xplore