Header logo is

PILCO: A Model-Based and Data-Efficient Approach to Policy Search

2011

Conference Paper

ei


In this paper, we introduce PILCO, a practical, data-efficient model-based policy search method. PILCO reduces model bias, one of the key problems of model-based reinforcement learning, in a principled way. By learning a probabilistic dynamics model and explicitly incorporating model uncertainty into long-term planning, PILCO can cope with very little data and facilitates learning from scratch in only a few trials. Policy evaluation is performed in closed form using state-of-the-art approximate inference. Furthermore, policy gradients are computed analytically for policy improvement. We report unprecedented learning efficiency on challenging and high-dimensional control tasks.

Author(s): Deisenroth, MP. and Rasmussen, CE.
Book Title: Proceedings of the 28th International Conference on Machine Learning, ICML 2011
Pages: 465-472
Year: 2011
Day: 0
Editors: L Getoor and T Scheffer
Publisher: Omnipress

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Place: Bellevue, Washington, USA

Digital: 0

Links: Web

BibTex

@inproceedings{DeisenrothRT2011,
  title = {PILCO: A Model-Based and Data-Efficient Approach to Policy Search},
  author = {Deisenroth, MP. and Rasmussen, CE.},
  booktitle = {Proceedings of the 28th International Conference on Machine Learning, ICML 2011},
  pages = {465-472},
  editors = {L Getoor and T Scheffer},
  publisher = {Omnipress},
  year = {2011},
  doi = {}
}