Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Data-driven approximate dynamic programming: A linear programming approach

Author(s):

T. Sutter, A. Kamoutsi, P. Mohajerin Esfahani, J. Lygeros
Conference/Journal:

IEEE Conference on Decision and Control
Abstract:

This article presents an approximation scheme for the infinite-dimensional linear programming formulation of discrete-time Markov control processes via a finite-dimensional convex program, when the dynamics are unknown and learned from data. We derive a probabilistic explicit error bound between the data-driven finite convex program and the orig- inal infinite linear program. We further discuss the sample complexity of the error bound which translates to the number of samples required for an a priori approximation accuracy. Our analysis sheds light on the impact of the choice of basis functions for approximating the true value function. Finally, the relevance of the method is illustrated on a truncated LQG problem.

Year:

2017
Type of Publication:

(01)Article
Supervisor:



File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@InProceedings { SutEtal:2017:IFA_5678,
    author={T. Sutter and A. Kamoutsi and P. Mohajerin Esfahani and J. Lygeros},
    title={{Data-driven approximate dynamic programming: A linear
	  programming approach}},
    booktitle={IEEE Conference on Decision and Control},
    pages={},
    year={2017},
    address={},
    month=dec,
    url={http://control.ee.ethz.ch/index.cgi?page=publications;action=details;id=5678}
}
Permanent link