AUGMENTED POLICY GRADIENT METHODS FOR EFFICIENT REINFORCEMENT LEARNING

no code implementations • 25 Sep 2019 • Kai Lagemann, Gregor Roering, Christoph Henke, Rene Vossen, Frank Hees

The influence of the ensemble of dynamics models on the policy update is controlled by adjusting the number of virtually performed rollouts in the next iteration according to the ratio of the real and virtual total reward.

Policy Gradient Methods reinforcement-learning +2

Learning Robust Manipulation Skills with Guided Policy Search via Generative Motor Reflexes

no code implementations • 15 Sep 2018 • Philipp Ennen, Pia Bresenitz, Rene Vossen, Frank Hees

However, due to the small number of real-world trajectory samples in Guided Policy Search, the resulting neural networks are only robust in the neighbourhood of the trajectory distribution explored by real-world interactions.