Verwandte Artikel zu Hierarchical Relative Entropy Policy Search: An Information...
Hierarchical Relative Entropy Policy Search: An Information Theoretic Learning Algorithm in Multimodal Solution Spaces for Real Robots - Softcover
Inhaltsangabe
Many real-world problems are inherently hierarchically structured. The use of this structure in an agent’s policy may well be the key to improved scalability and higher performance on motor skill tasks. However, such hierarchical structures cannot be exploited by current policy search algorithms. We concentrate on a basic, but highly relevant hierarchy — the `mixed option’ policy. Here, a gating network first decides which of the options to execute and, subsequently, the option-policy determines the action. Using a hierarchical setup for our learning method allows us to learn not only one solution to a problem but many. We base our algorithm on a recently proposed information theoretic policy search method, which addresses the exploitation-exploration trade-off by limiting the loss of information between policy updates.
Die Inhaltsangabe kann sich auf eine andere Ausgabe dieses Titels beziehen.
Reseña del editor
Many real-world problems are inherently hierarchically structured. The use of this structure in an agent's policy may well be the key to improved scalability and higher performance on motor skill tasks. However, such hierarchical structures cannot be exploited by current policy search algorithms. We concentrate on a basic, but highly relevant hierarchy - the `mixed option' policy. Here, a gating network first decides which of the options to execute and, subsequently, the option-policy determines the action. Using a hierarchical setup for our learning method allows us to learn not only one solution to a problem but many. We base our algorithm on a recently proposed information theoretic policy search method, which addresses the exploitation-exploration trade-off by limiting the loss of information between policy updates.
Biografía del autor
Christian Daniel studied computational engineering at Technische Universitaet Darmstadt and EPFL Lausanne and is pursuing a PhD in Robot Learning. His research focuses on developing new learning algorithms for autonomous robots, especially in the field of robot skill learning and hierarchical reinforcement learning.
„Über diesen Titel“ kann sich auf eine andere Ausgabe dieses Titels beziehen.
Neu kaufen
Diesen Artikel anzeigen
EUR 31,95
Gratis für den Versand innerhalb von/der Deutschland
Versandziele, Kosten & DauerSuchergebnisse für Hierarchical Relative Entropy Policy Search: An Information...
Foto des Verkäufers
Hierarchical Relative Entropy Policy Search
Verlag:
AV Akademikerverlag Jan 2014, 2014
ISBN 10: 3639475992
ISBN 13: 9783639475999
Neu
Taschenbuch
Anbieter: buchversandmimpf2000, Emtmannsberg, BAYE, Deutschland
Verkäuferbewertung 5 von 5 Sternen
Taschenbuch. Zustand: Neu. Neuware -Many real-world problems are inherently hierarchically structured. The use of this structure in an agent¿s policy may well be the key to improved scalability and higher performance on motor skill tasks. However, such hierarchical structures cannot be exploited by current policy search algorithms. We concentrate on a basic, but highly relevant hierarchy ¿ the `mixed option¿ policy. Here, a gating network first decides which of the options to execute and, subsequently, the option-policy determines the action. Using a hierarchical setup for our learning method allows us to learn not only one solution to a problem but many. We base our algorithm on a recently proposed information theoretic policy search method, which addresses the exploitation-exploration trade-off by limiting the loss of information between policy updates.VDM Verlag, Dudweiler Landstraße 99, 66123 Saarbrücken 68 pp. Englisch. Artikel-Nr. 9783639475999
Anzahl: 2 verfügbar