Improving Reinforcement Learning with Interactive Feedback and Affordances

Francisco Cruz , Sven Magg , Cornelius Weber , Stefan Wermter

IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EpiRob), pages 125--130, doi: 10.1109/DEVLRN.2014.6982975 - Oct 2014

Associated documents :

Interactive reinforcement learning constitutes an alternative for improving convergence speed in reinforcement learning methods. In this work, we investigate inter-agent training and present an approach for knowledge transfer in a domestic scenario where a first agent is trained by reinforcement learning and afterwards transfers selected knowledge to a second agent by instructions to achieve more efficient training. We combine this approach with action-space pruning by using knowledge on affordances and show that it significantly improves convergence speed in both classic and interactive reinforcement learning scenarios.

@InProceedings{CMWW14,
 	 author =  {Cruz, Francisco and Magg, Sven and Weber, Cornelius and Wermter, Stefan},
 	 title = {Improving Reinforcement Learning with Interactive Feedback and Affordances},
 	 booktitle = {IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EpiRob)},
 	 journal = {None},
 	 editors = {}
 	 number = {}
 	 volume = {}
 	 pages = {125--130},
 	 year = {2014},
 	 month = {Oct},
 	 publisher = {IEEE},
 	 doi = {10.1109/DEVLRN.2014.6982975},
 }