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

Research output: Contribution to journalArticleResearchpeer-review

Standard

Learned graphical models for probabilistic planning provide a new class of movement primitives. / Rückert, Elmar A.; Neumann, Gerhard; Toussaint, Marc et al.
In: Frontiers in computational neuroscience, Vol. 6.2013, No. January, 97, 02.01.2013.

Research output: Contribution to journalArticleResearchpeer-review

Vancouver

Rückert EA, Neumann G, Toussaint M, Maass W. Learned graphical models for probabilistic planning provide a new class of movement primitives. Frontiers in computational neuroscience. 2013 Jan 2;6.2013(January):97. Epub 2013 Jan 2. doi: 10.3389/fncom.2012.00097

Bibtex - Download

@article{d5a1aa1170fd4c008780a70a780a1b2b,
title = "Learned graphical models for probabilistic planning provide a new class of movement primitives",
abstract = "Biologicalmovementgenerationcombinesthreeinterestingaspects:itsmodularorganizationinmovementprimitives(MPs),itscharacteristicsofstochasticoptimalityunderperturbations,anditsefficiencyintermsoflearning.Acommonapproachtomotorskilllearningistoendowtheprimitiveswithdynamicalsystems.Here,theparametersoftheprimitiveindirectlydefinetheshapeofareferencetrajectory.WeproposeanalternativeMPrepresentationbasedonprobabilisticinferenceinlearnedgraphicalmodelswithnewandinterestingpropertiesthatcomplieswithsalientfeaturesofbiologicalmovementcontrol.Insteadofendowingtheprimitiveswithdynamicalsystems,weproposetoendowMPswithanintrinsicprobabilisticplanningsystem,integratingthepowerofstochasticoptimalcontrol(SOC)methodswithinaMP.Theparameterizationoftheprimitiveisagraphicalmodelthatrepresentsthedynamicsandintrinsiccostfunctionsuchthatinferenceinthisgraphicalmodelyieldsthecontrolpolicy.Weparameterizetheintrinsiccostfunctionusingtask-relevantfeatures,suchastheimportanceofpassingthroughcertainvia-points.Thesystemdynamicsaswellasintrinsiccostfunctionparametersarelearnedinareinforcementlearning(RL)setting.Weevaluateourapproachonacomplex4-linkbalancingtask.Ourexperimentsshowthatourmovementrepresentationfacilitateslearningsignificantlyandleadstobettergeneralizationtonewtasksettingswithoutre-learning.",
author = "R{\"u}ckert, {Elmar A.} and Gerhard Neumann and Marc Toussaint and Wolfgang Maass",
year = "2013",
month = jan,
day = "2",
doi = "10.3389/fncom.2012.00097",
language = "English",
volume = "6.2013",
journal = "Frontiers in computational neuroscience",
issn = "1662-5188",
number = "January",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Learned graphical models for probabilistic planning provide a new class of movement primitives

AU - Rückert, Elmar A.

AU - Neumann, Gerhard

AU - Toussaint, Marc

AU - Maass, Wolfgang

PY - 2013/1/2

Y1 - 2013/1/2

N2 - Biologicalmovementgenerationcombinesthreeinterestingaspects:itsmodularorganizationinmovementprimitives(MPs),itscharacteristicsofstochasticoptimalityunderperturbations,anditsefficiencyintermsoflearning.Acommonapproachtomotorskilllearningistoendowtheprimitiveswithdynamicalsystems.Here,theparametersoftheprimitiveindirectlydefinetheshapeofareferencetrajectory.WeproposeanalternativeMPrepresentationbasedonprobabilisticinferenceinlearnedgraphicalmodelswithnewandinterestingpropertiesthatcomplieswithsalientfeaturesofbiologicalmovementcontrol.Insteadofendowingtheprimitiveswithdynamicalsystems,weproposetoendowMPswithanintrinsicprobabilisticplanningsystem,integratingthepowerofstochasticoptimalcontrol(SOC)methodswithinaMP.Theparameterizationoftheprimitiveisagraphicalmodelthatrepresentsthedynamicsandintrinsiccostfunctionsuchthatinferenceinthisgraphicalmodelyieldsthecontrolpolicy.Weparameterizetheintrinsiccostfunctionusingtask-relevantfeatures,suchastheimportanceofpassingthroughcertainvia-points.Thesystemdynamicsaswellasintrinsiccostfunctionparametersarelearnedinareinforcementlearning(RL)setting.Weevaluateourapproachonacomplex4-linkbalancingtask.Ourexperimentsshowthatourmovementrepresentationfacilitateslearningsignificantlyandleadstobettergeneralizationtonewtasksettingswithoutre-learning.

AB - Biologicalmovementgenerationcombinesthreeinterestingaspects:itsmodularorganizationinmovementprimitives(MPs),itscharacteristicsofstochasticoptimalityunderperturbations,anditsefficiencyintermsoflearning.Acommonapproachtomotorskilllearningistoendowtheprimitiveswithdynamicalsystems.Here,theparametersoftheprimitiveindirectlydefinetheshapeofareferencetrajectory.WeproposeanalternativeMPrepresentationbasedonprobabilisticinferenceinlearnedgraphicalmodelswithnewandinterestingpropertiesthatcomplieswithsalientfeaturesofbiologicalmovementcontrol.Insteadofendowingtheprimitiveswithdynamicalsystems,weproposetoendowMPswithanintrinsicprobabilisticplanningsystem,integratingthepowerofstochasticoptimalcontrol(SOC)methodswithinaMP.Theparameterizationoftheprimitiveisagraphicalmodelthatrepresentsthedynamicsandintrinsiccostfunctionsuchthatinferenceinthisgraphicalmodelyieldsthecontrolpolicy.Weparameterizetheintrinsiccostfunctionusingtask-relevantfeatures,suchastheimportanceofpassingthroughcertainvia-points.Thesystemdynamicsaswellasintrinsiccostfunctionparametersarelearnedinareinforcementlearning(RL)setting.Weevaluateourapproachonacomplex4-linkbalancingtask.Ourexperimentsshowthatourmovementrepresentationfacilitateslearningsignificantlyandleadstobettergeneralizationtonewtasksettingswithoutre-learning.

U2 - 10.3389/fncom.2012.00097

DO - 10.3389/fncom.2012.00097

M3 - Article

VL - 6.2013

JO - Frontiers in computational neuroscience

JF - Frontiers in computational neuroscience

SN - 1662-5188

IS - January

M1 - 97

ER -