Deutsch
 
Hilfe Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

 
 
DownloadE-Mail
  Relative Entropy Inverse Reinforcement Learning

Boularias, A., Kober, J., & Peters, J. (2011). Relative Entropy Inverse Reinforcement Learning. In G. Gordon, D. Dunson, & M. Dudik (Eds.), JMLR Workshop and Conference Proceedings (pp. 182-189). Cambridge, MA, USA: MIT Press.

Item is

Externe Referenzen

einblenden:
ausblenden:
Beschreibung:
-
OA-Status:

Urheber

einblenden:
ausblenden:
 Urheber:
Boularias, A1, Autor           
Kober, J1, Autor           
Peters, J1, Autor           
Affiliations:
1Dept. Empirical Inference, Max Planck Institute for Intelligent Systems, Max Planck Society, ou_1497647              

Inhalt

einblenden:
ausblenden:
Schlagwörter: -
 Zusammenfassung: We consider the problem of imitation learning where the examples, demonstrated by an expert, cover only a small part of a large state space. Inverse Reinforcement Learning (IRL) provides an efficient tool for generalizing the demonstration, based on the assumption that the expert is optimally acting in a Markov Decision Process (MDP). Most of the past work on IRL requires that a (near)-optimal policy can be computed for different reward functions. However, this requirement can hardly be satisfied in systems with a large, or continuous, state space. In this paper, we propose a model-free IRL algorithm, where the relative entropy between the empirical distribution of the state-action trajectories under a uniform policy and their distribution under the learned policy is minimized by stochastic gradient descent. We compare this new approach to well-known IRL algorithms using approximate MDP models. Empirical results on simulated car racing, gridworld and ball-in-a-cup problems show that our approach is able to learn good policies from a small number of demonstrations.

Details

einblenden:
ausblenden:
Sprache(n):
 Datum: 2011-04
 Publikationsstatus: Erschienen
 Seiten: -
 Ort, Verlag, Ausgabe: -
 Inhaltsverzeichnis: -
 Art der Begutachtung: -
 Identifikatoren: URI: http://jmlr.csail.mit.edu/proceedings/papers/v15/boularias11a.html
BibTex Citekey: BoulariasKP2011
 Art des Abschluß: -

Veranstaltung

einblenden:
ausblenden:
Titel: Fourteenth International Conference on Artificial Intelligence and Statistics (AISTATS 2011)
Veranstaltungsort: Ft. Lauderdale, FL, USA
Start-/Enddatum: 2011-04-11 - 2011-04-13

Entscheidung

einblenden:

Projektinformation

einblenden:

Quelle 1

einblenden:
ausblenden:
Titel: JMLR Workshop and Conference Proceedings
Genre der Quelle: Konferenzband
 Urheber:
Gordon, G, Herausgeber
Dunson, D, Herausgeber
Dudik, M, Herausgeber
Affiliations:
-
Ort, Verlag, Ausgabe: Cambridge, MA, USA : MIT Press
Seiten: - Band / Heft: 15 Artikelnummer: - Start- / Endseite: 182 - 189 Identifikator: -