Deutsch
 
Hilfe Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

 
 
DownloadE-Mail
  Recurrent Policy Gradients

Wierstra, D., Förster A, Peters, J., & Schmidhuber, J. (2010). Recurrent Policy Gradients. Logic Journal of the IGPL, 18(5), 620-634. doi:10.1093/jigpal/jzp049.

Item is

Externe Referenzen

einblenden:

Urheber

einblenden:
ausblenden:
 Urheber:
Wierstra, D, Autor
Förster A, Peters, J1, 2, Autor           
Schmidhuber, J, Autor
Affiliations:
1Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497795              
2Dept. Empirical Inference, Max Planck Institute for Intelligent Systems, Max Planck Society, ou_1497647              

Inhalt

einblenden:
ausblenden:
Schlagwörter: -
 Zusammenfassung: Reinforcement learning for partially observable Markov decision problems (POMDPs) is a challenge as it requires policies with an internal state. Traditional approaches suffer significantly from this shortcoming and usually make strong assumptions on the problem domain such as perfect system models, state-estimators and a Markovian hidden system. Recurrent neural networks (RNNs) offer a natural framework for dealing with policy learning using hidden state and require only few limiting assumptions. As they can be trained well using gradient descent, they are suited for policy gradient approaches. In this paper, we present a policy gradient method, the Recurrent Policy Gradient which constitutes a model-free reinforcement learning method. It is aimed at training limited-memory stochastic policies on problems which require long-term memories of past observations. The approach involves approximating a policy gradient for a recurrent neural network by backpropagating return-weighted characteristic eligibilities through time. Using a ‘‘Long Short-Term Memory’’ RNN architecture, we are able to outperform previous RL methods on three important benchmark tasks. Furthermore, we show that using history-dependent baselines helps reducing estimation variance significantly, thus enabling our approach to tackle more challenging, highly stochastic environments.

Details

einblenden:
ausblenden:
Sprache(n):
 Datum: 2010-10
 Publikationsstatus: Erschienen
 Seiten: -
 Ort, Verlag, Ausgabe: -
 Inhaltsverzeichnis: -
 Art der Begutachtung: -
 Identifikatoren: URI: http://jigpal.oxfordjournals.org/cgi/reprint/jzp049v1
DOI: 10.1093/jigpal/jzp049
BibTex Citekey: 5879
 Art des Abschluß: -

Veranstaltung

einblenden:

Entscheidung

einblenden:

Projektinformation

einblenden:

Quelle 1

einblenden:
ausblenden:
Titel: Logic Journal of the IGPL
Genre der Quelle: Zeitschrift
 Urheber:
Affiliations:
Ort, Verlag, Ausgabe: -
Seiten: - Band / Heft: 18 (5) Artikelnummer: - Start- / Endseite: 620 - 634 Identifikator: -