… is to show that demonstrations can be used with reinforcement learning to solve complex …
demonstrations, a method of annealing away the effect of the demonstrations when the learned …

… supervised and reinforcement losses. This makes NAC robust to suboptimal demonstration
data, since … We show that our unified reinforcement learning algorithm can learn robustly and …

… Enabling learning in this framework opens up the possibility of applying RL to many real world
problems where demonstration data … We propose a new deep reinforcement learning al- …

… provides a significant step toward practical realworld learning of complex dexterous
manipulation. In future work, we hope to learn policies on real hardware systems, further reduce …

… from the creation of the demonstration data set, to learning methods from demonstrations,
and optimization by combining demonstration learning with different machine learning methods…

… trend is to learn control policies in simulation and then transfer them over to the real world.
To date, no work has explored whether it is possible to learn and transfer deformable object …

… lies on reinforcement learning and how learning from demonstration can be beneficial in
this context. We divide reinforcement learning into two categories: reinforcement learning for …

… While in this work, training was purely conducted in simulation, in future work we will
investigate how real-world human demonstrations can be leveraged and how this navigation …

… For providing demonstration for the real world tasks we used the same setup, this time
controlling a second robotic arm. Separating the arm we were controlling and the arm which …

… maker can learn about … study, we propose a new reinforcement learning algorithm for the
decision maker called Relational Exploration with Demonstrations (REX-D), which can learn to …