Perceiving Systems, Computer Vision

Model-Based Reinforcement Learning with Continuous States and Actions

2008

Conference Paper

ei


Finding an optimal policy in a reinforcement learning (RL) framework with continuous state and action spaces is challenging. Approximate solutions are often inevitable. GPDP is an approximate dynamic programming algorithm based on Gaussian process (GP) models for the value functions. In this paper, we extend GPDP to the case of unknown transition dynamics. After building a GP model for the transition dynamics, we apply GPDP to this model and determine a continuous-valued policy in the entire state space. We apply the resulting controller to the underpowered pendulum swing up. Moreover, we compare our results on this RL task to a nearly optimal discrete DP solution in a fully known environment.

Author(s): Deisenroth, MP. and Rasmussen, CE. and Peters, J.
Book Title: ESANN 2008
Journal: Advances in Computational Intelligence and Learning: Proceedings of the European Symposium on Artificial Neural Networks (ESANN 2008)
Pages: 19-24
Year: 2008
Month: April
Day: 0
Editors: Verleysen, M.
Publisher: d-side

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

Event Name: European Symposium on Artificial Neural Networks
Event Place: Bruges, Belgium

Address: Evere, Belgium
Digital: 0
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{4977,
  title = {Model-Based Reinforcement Learning with Continuous States and Actions},
  author = {Deisenroth, MP. and Rasmussen, CE. and Peters, J.},
  journal = {Advances in Computational Intelligence and Learning: Proceedings of the European Symposium on Artificial Neural Networks (ESANN 2008)},
  booktitle = {ESANN 2008},
  pages = {19-24},
  editors = {Verleysen, M. },
  publisher = {d-side},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Evere, Belgium},
  month = apr,
  year = {2008},
  doi = {},
  month_numeric = {4}
}