Low Dimensional State Representation Learning with Reward-shaped Priors

Nicolò Botteghi, Ruben Obbink, Daan Geijs, Mannes Poel, Beril Sirmacek, Christoph Brune, Abeje Mersha, Stefano Stramigioli

Research output: Contribution to conferencePaperAcademicpeer-review

3 Downloads (Pure)

Abstract

Reinforcement Learning has been able to solve many complicated robotics tasks without any need for feature engineering in an end-to-end fashion. However, learning the optimal policy directly from the sensory inputs, i.e the observations, often requires processing and storage of a huge amount of data. In the context of robotics, the cost of data from real robotics hardware is usually very high, thus solutions that achieve high sample-efficiency are needed. We propose a method that aims at learning a mapping from the observations into a lower-dimensional state space. This mapping is learned with unsupervised learning using loss functions shaped to incorporate prior knowledge of the environment and the task. Using the samples from the state space, the optimal policy is quickly and efficiently learned. We test the method on several mobile robot navigation tasks in a simulation environment and also on a real robot.
Original languageEnglish
Number of pages8
Publication statusPublished - 12 Jan 2021
Event25th International Conference on Pattern Recognition, ICPR 2020 - Online conference
Duration: 12 Jan 202115 Jan 2021
Conference number: 25
https://www.micc.unifi.it/icpr2020/

Conference

Conference25th International Conference on Pattern Recognition, ICPR 2020
Abbreviated titleICPR
Period12/01/2115/01/21
Internet address

Keywords

  • Deep learning
  • Reinforcement learning
  • State representation learning
  • Robotics

Fingerprint Dive into the research topics of 'Low Dimensional State Representation Learning with Reward-shaped Priors'. Together they form a unique fingerprint.

Cite this