Learned graphical models for probabilistic planning provide a new class of movement primitives

Learned graphical models for probabilistic planning provide a new class of movement primitives

Biological movement generation combines three interesting aspects: itsmodular organization in movement primitives, its characteristics ofstochastic optimality under perturbations, and its efficiency in termsof learning. A common approach to motor skill learning is toendow the primitives with dynamic...

Full description

Bibliographic Details
Main Authors: Elmar A. Rückert, Gerhard eNeumann, Marc eToussaint, Wolfgang eMaass
Format: Article
Language:English
Published: Frontiers Media S.A. 2013-01-01
Series:Frontiers in Computational Neuroscience
Subjects:
reinforcement learning
Graphical Models
Motor planning
movement primitives
optimal control
Online Access:http://journal.frontiersin.org/Journal/10.3389/fncom.2012.00097/full

Internet

http://journal.frontiersin.org/Journal/10.3389/fncom.2012.00097/full

Similar Items