1 paper
Daniel Palenicek, Michael Lutter, João Carvalho +3
Model-based reinforcement learning aims to increase sample efficiency, but the accuracy of dynamics models and the resulting compounding errors are often seen as key limitations. T…