Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors

Humans intuitively solve tasks in versatile ways, varying their behavior in terms of trajectory-based planning and for individual steps. Thus, they can easily generalize and adapt to new and changing environments. Current Imitation Learning algorithms often only consider unimodal expert demonstratio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2022-11
Hauptverfasser: Freymuth, Niklas, Schreiber, Nicolas, Becker, Philipp, Taranovic, Aleksandar, Neumann, Gerhard
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Freymuth, Niklas
Schreiber, Nicolas
Becker, Philipp
Taranovic, Aleksandar
Neumann, Gerhard
description Humans intuitively solve tasks in versatile ways, varying their behavior in terms of trajectory-based planning and for individual steps. Thus, they can easily generalize and adapt to new and changing environments. Current Imitation Learning algorithms often only consider unimodal expert demonstrations and act in a state-action-based setting, making it difficult for them to imitate human behavior in case of versatile demonstrations. Instead, we combine a mixture of movement primitives with a distribution matching objective to learn versatile behaviors that match the expert's behavior and versatility. To facilitate generalization to novel task configurations, we do not directly match the agent's and expert's trajectory distributions but rather work with concise geometric descriptors which generalize well to unseen task configurations. We empirically validate our method on various robot tasks using versatile human demonstrations and compare to imitation learning algorithms in a state-action setting as well as a trajectory-based setting. We find that the geometric descriptors greatly help in generalizing to new task configurations and that combining them with our distribution-matching objective is crucial for representing and reproducing versatile behavior.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2725745092</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2725745092</sourcerecordid><originalsourceid>FETCH-proquest_journals_27257450923</originalsourceid><addsrcrecordid>eNqNjr0KwjAURoMgWLTvEHAuxKS1uvrv4CLiWmK5tSltUu-Ngm9vBB_A6QznfPANWCSVmiWLVMoRi4kaIYSc5zLLVMTOR1sBorF3fgUk7U0LfAW1fhmHvELX8Q10zpLH4AL57c1P2pf1d7IH14FHU4aISjS9d0gTNqx0SxD_OGbT3fayPiQ9uscTyBeNe6INqpDhRZ5mYinVf9UHSu5AdQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2725745092</pqid></control><display><type>article</type><title>Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors</title><source>Free E- Journals</source><creator>Freymuth, Niklas ; Schreiber, Nicolas ; Becker, Philipp ; Taranovic, Aleksandar ; Neumann, Gerhard</creator><creatorcontrib>Freymuth, Niklas ; Schreiber, Nicolas ; Becker, Philipp ; Taranovic, Aleksandar ; Neumann, Gerhard</creatorcontrib><description>Humans intuitively solve tasks in versatile ways, varying their behavior in terms of trajectory-based planning and for individual steps. Thus, they can easily generalize and adapt to new and changing environments. Current Imitation Learning algorithms often only consider unimodal expert demonstrations and act in a state-action-based setting, making it difficult for them to imitate human behavior in case of versatile demonstrations. Instead, we combine a mixture of movement primitives with a distribution matching objective to learn versatile behaviors that match the expert's behavior and versatility. To facilitate generalization to novel task configurations, we do not directly match the agent's and expert's trajectory distributions but rather work with concise geometric descriptors which generalize well to unseen task configurations. We empirically validate our method on various robot tasks using versatile human demonstrations and compare to imitation learning algorithms in a state-action setting as well as a trajectory-based setting. We find that the geometric descriptors greatly help in generalizing to new task configurations and that combining them with our distribution-matching objective is crucial for representing and reproducing versatile behavior.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Changing environments ; Configurations ; Human behavior ; Machine learning ; Matching ; Trajectory planning</subject><ispartof>arXiv.org, 2022-11</ispartof><rights>2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Freymuth, Niklas</creatorcontrib><creatorcontrib>Schreiber, Nicolas</creatorcontrib><creatorcontrib>Becker, Philipp</creatorcontrib><creatorcontrib>Taranovic, Aleksandar</creatorcontrib><creatorcontrib>Neumann, Gerhard</creatorcontrib><title>Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors</title><title>arXiv.org</title><description>Humans intuitively solve tasks in versatile ways, varying their behavior in terms of trajectory-based planning and for individual steps. Thus, they can easily generalize and adapt to new and changing environments. Current Imitation Learning algorithms often only consider unimodal expert demonstrations and act in a state-action-based setting, making it difficult for them to imitate human behavior in case of versatile demonstrations. Instead, we combine a mixture of movement primitives with a distribution matching objective to learn versatile behaviors that match the expert's behavior and versatility. To facilitate generalization to novel task configurations, we do not directly match the agent's and expert's trajectory distributions but rather work with concise geometric descriptors which generalize well to unseen task configurations. We empirically validate our method on various robot tasks using versatile human demonstrations and compare to imitation learning algorithms in a state-action setting as well as a trajectory-based setting. We find that the geometric descriptors greatly help in generalizing to new task configurations and that combining them with our distribution-matching objective is crucial for representing and reproducing versatile behavior.</description><subject>Algorithms</subject><subject>Changing environments</subject><subject>Configurations</subject><subject>Human behavior</subject><subject>Machine learning</subject><subject>Matching</subject><subject>Trajectory planning</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNjr0KwjAURoMgWLTvEHAuxKS1uvrv4CLiWmK5tSltUu-Ngm9vBB_A6QznfPANWCSVmiWLVMoRi4kaIYSc5zLLVMTOR1sBorF3fgUk7U0LfAW1fhmHvELX8Q10zpLH4AL57c1P2pf1d7IH14FHU4aISjS9d0gTNqx0SxD_OGbT3fayPiQ9uscTyBeNe6INqpDhRZ5mYinVf9UHSu5AdQ</recordid><startdate>20221109</startdate><enddate>20221109</enddate><creator>Freymuth, Niklas</creator><creator>Schreiber, Nicolas</creator><creator>Becker, Philipp</creator><creator>Taranovic, Aleksandar</creator><creator>Neumann, Gerhard</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20221109</creationdate><title>Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors</title><author>Freymuth, Niklas ; Schreiber, Nicolas ; Becker, Philipp ; Taranovic, Aleksandar ; Neumann, Gerhard</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_27257450923</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Algorithms</topic><topic>Changing environments</topic><topic>Configurations</topic><topic>Human behavior</topic><topic>Machine learning</topic><topic>Matching</topic><topic>Trajectory planning</topic><toplevel>online_resources</toplevel><creatorcontrib>Freymuth, Niklas</creatorcontrib><creatorcontrib>Schreiber, Nicolas</creatorcontrib><creatorcontrib>Becker, Philipp</creatorcontrib><creatorcontrib>Taranovic, Aleksandar</creatorcontrib><creatorcontrib>Neumann, Gerhard</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Freymuth, Niklas</au><au>Schreiber, Nicolas</au><au>Becker, Philipp</au><au>Taranovic, Aleksandar</au><au>Neumann, Gerhard</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors</atitle><jtitle>arXiv.org</jtitle><date>2022-11-09</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>Humans intuitively solve tasks in versatile ways, varying their behavior in terms of trajectory-based planning and for individual steps. Thus, they can easily generalize and adapt to new and changing environments. Current Imitation Learning algorithms often only consider unimodal expert demonstrations and act in a state-action-based setting, making it difficult for them to imitate human behavior in case of versatile demonstrations. Instead, we combine a mixture of movement primitives with a distribution matching objective to learn versatile behaviors that match the expert's behavior and versatility. To facilitate generalization to novel task configurations, we do not directly match the agent's and expert's trajectory distributions but rather work with concise geometric descriptors which generalize well to unseen task configurations. We empirically validate our method on various robot tasks using versatile human demonstrations and compare to imitation learning algorithms in a state-action setting as well as a trajectory-based setting. We find that the geometric descriptors greatly help in generalizing to new task configurations and that combining them with our distribution-matching objective is crucial for representing and reproducing versatile behavior.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2022-11
issn 2331-8422
language eng
recordid cdi_proquest_journals_2725745092
source Free E- Journals
subjects Algorithms
Changing environments
Configurations
Human behavior
Machine learning
Matching
Trajectory planning
title Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-21T18%3A26%3A24IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Inferring%20Versatile%20Behavior%20from%20Demonstrations%20by%20Matching%20Geometric%20Descriptors&rft.jtitle=arXiv.org&rft.au=Freymuth,%20Niklas&rft.date=2022-11-09&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2725745092%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2725745092&rft_id=info:pmid/&rfr_iscdi=true