Non-linear regression algorithms for motor skill acquisition: a comparison - Archive ouverte HAL Access content directly
Conference Papers Year :

Non-linear regression algorithms for motor skill acquisition: a comparison

Abstract

Endowing robots with the capability to learn is an important goal for the robotics re-search community. One part of this research is focused on learning skills, where usually two learning paradigms are used sequentially. First, a robot learns a motor primitive by demonstration (or imitation). Then, it improves this motor primitive with respect to some externally defined criterion. In this paper, we study how the representation used in the demonstration learning step can influence the performance of the policy improvement step. We provide a conceptual survey of different demonstration learning al-gorithms and perform an empirical comparison of their performance when combined with a subsequent policy improvement step.
Fichier principal
Vignette du fichier
jfpda2014.pdf (841.48 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01090848 , version 1 (04-12-2014)

Identifiers

  • HAL Id : hal-01090848 , version 1

Cite

Thibaut Munzer, Freek Stulp, Olivier Sigaud. Non-linear regression algorithms for motor skill acquisition: a comparison. 9èmes Journées Francophones de Planification, Décision et Apprentissage, May 2014, Liège, Belgium. ⟨hal-01090848⟩
193 View
289 Download

Share

Gmail Facebook Twitter LinkedIn More