Learning Policies for Markov Decision Processes in Continuous Spaces | IEEE Conference Publication | IEEE Xplore