ResearchSpace

Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning

Show simple item record

dc.contributor.author Ranchod, P
dc.contributor.author Rosman, Benjamin S
dc.contributor.author Konidaris, G
dc.date.accessioned 2015-11-16T07:36:29Z
dc.date.available 2015-11-16T07:36:29Z
dc.date.issued 2015-10
dc.identifier.citation Ranchod, P, Rosman, B.S. and Konidaris, G. 2015. Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning. In: IEEE/RSJ International Conference on Intelligent Robots and Systems, Hamburg Germany, September-October 2015 en_US
dc.identifier.uri http://irl.cs.duke.edu/pubs/npbrs.pdf
dc.identifier.uri http://hdl.handle.net/10204/8290
dc.description IEEE/RSJ International Conference on Intelligent Robots and Systems, Hamburg Germany, September-October 2015. en_US
dc.description.abstract We present a method for segmenting a set of unstructured demonstration trajectories to discover reusable skills using inverse reinforcement learning (IRL). Each skill is characterised by a latent reward function which the demonstrator is assumed to be optimizing. The skill boundaries and the number of skills making up each demonstration are unknown. We use a Bayesian nonparametric approach to propose skill segmentations and maximum entropy inverse reinforcement learning to infer reward functions from the segments. This method produces a set of Markov Decision Processes (MDPs) that best describe the input trajectories. We evaluate this approach in a car driving domain and a simulated quadcopter obstacle course, showing that it is able to recover demonstrated skills more effectively than existing methods. en_US
dc.language.iso en en_US
dc.publisher IEEE en_US
dc.relation.ispartofseries Workflow;15679
dc.subject Inverse reinforcement learning en_US
dc.subject Nonparametric bayesian methods en_US
dc.subject Skill discovery en_US
dc.subject Imitation learning en_US
dc.title Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning en_US
dc.type Conference Presentation en_US
dc.identifier.apacitation Ranchod, P., Rosman, B. S., & Konidaris, G. (2015). Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning. IEEE. http://hdl.handle.net/10204/8290 en_ZA
dc.identifier.chicagocitation Ranchod, P, Benjamin S Rosman, and G Konidaris. "Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning." (2015): http://hdl.handle.net/10204/8290 en_ZA
dc.identifier.vancouvercitation Ranchod P, Rosman BS, Konidaris G, Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning; IEEE; 2015. http://hdl.handle.net/10204/8290 . en_ZA
dc.identifier.ris TY - Conference Presentation AU - Ranchod, P AU - Rosman, Benjamin S AU - Konidaris, G AB - We present a method for segmenting a set of unstructured demonstration trajectories to discover reusable skills using inverse reinforcement learning (IRL). Each skill is characterised by a latent reward function which the demonstrator is assumed to be optimizing. The skill boundaries and the number of skills making up each demonstration are unknown. We use a Bayesian nonparametric approach to propose skill segmentations and maximum entropy inverse reinforcement learning to infer reward functions from the segments. This method produces a set of Markov Decision Processes (MDPs) that best describe the input trajectories. We evaluate this approach in a car driving domain and a simulated quadcopter obstacle course, showing that it is able to recover demonstrated skills more effectively than existing methods. DA - 2015-10 DB - ResearchSpace DP - CSIR KW - Inverse reinforcement learning KW - Nonparametric bayesian methods KW - Skill discovery KW - Imitation learning LK - https://researchspace.csir.co.za PY - 2015 T1 - Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning TI - Nonparametric bayesian reward segmentation for skill discovery using inverse reinforcement learning UR - http://hdl.handle.net/10204/8290 ER - en_ZA


Files in this item

This item appears in the following Collection(s)

Show simple item record