Bridging Scenarios in Reinforcement Learning with Continuously Generated Relaying Predictive Models | IEEE Conference Publication | IEEE Xplore