Publication Details

SELECT * FROM publications WHERE Record_Number=2675
Reference TypeConference Proceedings
Author(s)Peters, J.;Schaal, S.
Year2007
TitleReinforcement learning by reward-weighted regression for operational space control
Journal/Conference/Book TitleProceedings of the International Conference on Machine Learning (ICML2007)
Keywordsreinforcement learning, operational space control, weighted regression
AbstractMany robot control problems of practical importance, including operational space control, can be reformulated as immediate reward reinforcement learning problems. However, few of the known optimization or reinforcement learning algorithms can be used in online learning control for robots, as they are either prohibitively slow, do not scale to interesting domains of complex robots, or require trying out policies generated by random search, which are infeasible for a physical system. Using a generalization of the EM-base reinforcement learning framework suggested by Dayan & Hinton, we reduce the problem of learning with immediate rewards to a reward-weighted regression problem with an adaptive, integrated reward transformation for faster convergence. The resulting algorithm is efficient, learns smoothly without dangerous jumps in solution space, and works well in applications of complex high degree-of-freedom robots.
Place PublishedCorvallis, Oregon, June 19-21
Short TitleReinforcement learning by reward-weighted regression for operational space control
URL(s) http://www-clmc.usc.edu/publications//P/peters_ICML2007.pdf

  

zum Seitenanfang