Can I read Representation Matters: Improving Perception and Exploration for Robotics on EtoBox?

Representation Matters: Improving Perception and Exploration for Robotics by Wulfmeier, Markus; Byravan, Arunkumar; Hertweck, Tim; Higgins, Irina; Gupta, Ankush; Kulkarni, Tejas; Reynolds, Malcolm; Teplyashin, Denis; Hafner, Roland; Lampe, Thomas; Riedmiller, Martin is a scholarly article available to read on EtoBox.

What is Representation Matters: Improving Perception and Exploration for Robotics about?

Projecting high-dimensional environment observations into lower-dimensional structured representations can considerably improve data-efficiency for reinforcement learning in domains with limited data such as robotics. Can a single generally useful representation be found? In order to answer this question, it is important to understand how the representation will be used by the agent and what properties such a 'good' representation should have. In this paper we systematically evaluate a number of common learnt and hand-engineered representations in the context of three robotics tasks: lifting, stacking and pushing of 3D blocks. The representations are evaluated in two use-cases: as input to the agent, or as a source of auxiliary tasks. Furthermore, the value of each representation is evaluated in terms of three properties: dimensionality, observability and disentanglement. We can significantly improve performance in both use-cases and demonstrate that some representations can perform commensurate to simulator states as agent inputs. Finally, our results challenge common intuitions by demonstrating that: 1) dimensionality strongly matters for task generation, but is negligible for in

Author

Wulfmeier, Markus; Byravan, Arunkumar; Hertweck, Tim; Higgins, Irina; Gupta, Ankush; Kulkarni, Tejas; Reynolds, Malcolm; Teplyashin, Denis; Hafner, Roland; Lampe, Thomas; Riedmiller, Martin

Published

2020

Language

EN