Data-Efficient Generalization of Robot Skills with Contextual Policy Search

File Description SizeFormat 
Kupcsik_AAAI_2013.pdfAccepted version1.57 MBAdobe PDFView/Open
Title: Data-Efficient Generalization of Robot Skills with Contextual Policy Search
Authors: Kupcsik, A
Deisenroth, MP
Peters, J
Neumann, G
Item Type: Conference Paper
Abstract: In robotics, controllers make the robot solve a task within a specific context. The context can describe the objectives of the robot or physical properties of the environment and is always specified before task execution. To generalize the controller to multiple contexts, we follow a hierarchical approach for policy learning: A lower-level policy controls the robot for a given context and an upper-level policy generalizes among contexts. Current approaches for learning such upper-level policies are based on model-free policy search, which require an excessive number of interactions of the robot with its environment. More data-efficient policy search approaches are model based but, thus far, without the capability of learning hierarchical policies. We propose a new model-based policy search approach that can also learn contextual upper-level policies. Our approach is based on learning probabilistic forward models for long-term predictions. Using these predictions, we use information-theoretic insights to improve the upper-level policy. Our method achieves a substantial improvement in learning speed compared to existing methods on simulated and real robotic tasks. Copyright © 2013, Association for the Advancement of Artificial Intelligence ( All rights reserved.
Issue Date: 1-Dec-2013
ISBN: 9781577356158
Publisher: AAAI
Journal / Book Title: Proceedings of the AAAI Conference on Artificial Intelligence
Copyright Statement: © 2013 Association for the Advancement of Artificial Intelligence
Conference Name: 27th AAAI Conference
Notes: owner: marc timestamp: 2012.11.18
Publisher URL:
Start Date: 2013-07-14
Finish Date: 2013-07-18
Conference Place: Washington, USA
Appears in Collections:Computing

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx