Sample-Efficient Policy Search with a Trajectory Autoencoder
Alexander Fabisch, Frank Kirchner
In Proceedings of the 4th Robot Learning Workshop: Self-Supervised and Lifelong Learning, (NeurIPS-2021), 14.12.2021, virtuell, n.n., Dec/2021.
We introduce a trajectory generator that can be used to perform sample-efficient policy
search with Bayesian optimization (BO). BO is a sample-efficient approach to
direct policy search that usually does not scale well with the number of parameters.
Our trajectory generator is able to map a compact representation of trajectories to a
high-dimensional trajectory space so that BO can search in the low-dimensional
space. The trajectory generator will be trained as part of a variational autoencoder
on demonstrations from an expert. The trajectory generator contains a trajectory
layer, which is a new building block for neural networks that enforces smoothness
on generated trajectories. We evaluate our approach with grasping on a real robot.