Option Discovery in Hierarchical Reinforcement Learning using Spatio-Temporal Clustering

This paper introduces an automated skill acquisition framework in reinforcement learning which involves identifying a hierarchical description of the given task in terms of abstract states and extended actions between abstract states. Identifying such structures present in the task provides ways to...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2020-09
Hauptverfasser: Srinivas, Aravind, Krishnamurthy, Ramnandan, Kumar, Peeyush, Ravindran, Balaraman
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 Srinivas, Aravind
Krishnamurthy, Ramnandan
Kumar, Peeyush
Ravindran, Balaraman
description This paper introduces an automated skill acquisition framework in reinforcement learning which involves identifying a hierarchical description of the given task in terms of abstract states and extended actions between abstract states. Identifying such structures present in the task provides ways to simplify and speed up reinforcement learning algorithms. These structures also help to generalize such algorithms over multiple tasks without relearning policies from scratch. We use ideas from dynamical systems to find metastable regions in the state space and associate them with abstract states. The spectral clustering algorithm PCCA+ is used to identify suitable abstractions aligned to the underlying structure. Skills are defined in terms of the sequence of actions that lead to transitions between such abstract states. The connectivity information from PCCA+ is used to generate these skills or options. These skills are independent of the learning task and can be efficiently reused across a variety of tasks defined over the same model. This approach works well even without the exact model of the environment by using sample trajectories to construct an approximate estimate. We also present our approach to scaling the skill acquisition framework to complex tasks with large state spaces for which we perform state aggregation using the representation learned from an action conditional video prediction network and use the skill acquisition framework on the aggregated state space.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2080574930</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2080574930</sourcerecordid><originalsourceid>FETCH-proquest_journals_20805749303</originalsourceid><addsrcrecordid>eNqNi7EKwjAURYMgWLT_EHAuxKS1da5KB0HQDm4lhFdNaZP60gj-vRH8AJd7hnPujERciE1SpJwvSOxcxxjj25xnmYjI7TxO2hq6107ZF-CbakMrDShRPbSSPb2ANq1FBQOYiZ5AotHmTr377nWU4Z7UMIwWQ1z23k2AQa3IvJW9g_jHJVkfD3VZJSPapwc3NZ31aIJqOCtYlqc7wcR_1QfxYUKY</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2080574930</pqid></control><display><type>article</type><title>Option Discovery in Hierarchical Reinforcement Learning using Spatio-Temporal Clustering</title><source>Freely Accessible Journals</source><creator>Srinivas, Aravind ; Krishnamurthy, Ramnandan ; Kumar, Peeyush ; Ravindran, Balaraman</creator><creatorcontrib>Srinivas, Aravind ; Krishnamurthy, Ramnandan ; Kumar, Peeyush ; Ravindran, Balaraman</creatorcontrib><description>This paper introduces an automated skill acquisition framework in reinforcement learning which involves identifying a hierarchical description of the given task in terms of abstract states and extended actions between abstract states. Identifying such structures present in the task provides ways to simplify and speed up reinforcement learning algorithms. These structures also help to generalize such algorithms over multiple tasks without relearning policies from scratch. We use ideas from dynamical systems to find metastable regions in the state space and associate them with abstract states. The spectral clustering algorithm PCCA+ is used to identify suitable abstractions aligned to the underlying structure. Skills are defined in terms of the sequence of actions that lead to transitions between such abstract states. The connectivity information from PCCA+ is used to generate these skills or options. These skills are independent of the learning task and can be efficiently reused across a variety of tasks defined over the same model. This approach works well even without the exact model of the environment by using sample trajectories to construct an approximate estimate. We also present our approach to scaling the skill acquisition framework to complex tasks with large state spaces for which we perform state aggregation using the representation learned from an action conditional video prediction network and use the skill acquisition framework on the aggregated state space.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Clustering ; Machine learning ; Metastable region ; Skills ; Task complexity</subject><ispartof>arXiv.org, 2020-09</ispartof><rights>2020. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (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>Srinivas, Aravind</creatorcontrib><creatorcontrib>Krishnamurthy, Ramnandan</creatorcontrib><creatorcontrib>Kumar, Peeyush</creatorcontrib><creatorcontrib>Ravindran, Balaraman</creatorcontrib><title>Option Discovery in Hierarchical Reinforcement Learning using Spatio-Temporal Clustering</title><title>arXiv.org</title><description>This paper introduces an automated skill acquisition framework in reinforcement learning which involves identifying a hierarchical description of the given task in terms of abstract states and extended actions between abstract states. Identifying such structures present in the task provides ways to simplify and speed up reinforcement learning algorithms. These structures also help to generalize such algorithms over multiple tasks without relearning policies from scratch. We use ideas from dynamical systems to find metastable regions in the state space and associate them with abstract states. The spectral clustering algorithm PCCA+ is used to identify suitable abstractions aligned to the underlying structure. Skills are defined in terms of the sequence of actions that lead to transitions between such abstract states. The connectivity information from PCCA+ is used to generate these skills or options. These skills are independent of the learning task and can be efficiently reused across a variety of tasks defined over the same model. This approach works well even without the exact model of the environment by using sample trajectories to construct an approximate estimate. We also present our approach to scaling the skill acquisition framework to complex tasks with large state spaces for which we perform state aggregation using the representation learned from an action conditional video prediction network and use the skill acquisition framework on the aggregated state space.</description><subject>Algorithms</subject><subject>Clustering</subject><subject>Machine learning</subject><subject>Metastable region</subject><subject>Skills</subject><subject>Task complexity</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNi7EKwjAURYMgWLT_EHAuxKS1da5KB0HQDm4lhFdNaZP60gj-vRH8AJd7hnPujERciE1SpJwvSOxcxxjj25xnmYjI7TxO2hq6107ZF-CbakMrDShRPbSSPb2ANq1FBQOYiZ5AotHmTr377nWU4Z7UMIwWQ1z23k2AQa3IvJW9g_jHJVkfD3VZJSPapwc3NZ31aIJqOCtYlqc7wcR_1QfxYUKY</recordid><startdate>20200921</startdate><enddate>20200921</enddate><creator>Srinivas, Aravind</creator><creator>Krishnamurthy, Ramnandan</creator><creator>Kumar, Peeyush</creator><creator>Ravindran, Balaraman</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>20200921</creationdate><title>Option Discovery in Hierarchical Reinforcement Learning using Spatio-Temporal Clustering</title><author>Srinivas, Aravind ; Krishnamurthy, Ramnandan ; Kumar, Peeyush ; Ravindran, Balaraman</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_20805749303</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Algorithms</topic><topic>Clustering</topic><topic>Machine learning</topic><topic>Metastable region</topic><topic>Skills</topic><topic>Task complexity</topic><toplevel>online_resources</toplevel><creatorcontrib>Srinivas, Aravind</creatorcontrib><creatorcontrib>Krishnamurthy, Ramnandan</creatorcontrib><creatorcontrib>Kumar, Peeyush</creatorcontrib><creatorcontrib>Ravindran, Balaraman</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>Srinivas, Aravind</au><au>Krishnamurthy, Ramnandan</au><au>Kumar, Peeyush</au><au>Ravindran, Balaraman</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Option Discovery in Hierarchical Reinforcement Learning using Spatio-Temporal Clustering</atitle><jtitle>arXiv.org</jtitle><date>2020-09-21</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>This paper introduces an automated skill acquisition framework in reinforcement learning which involves identifying a hierarchical description of the given task in terms of abstract states and extended actions between abstract states. Identifying such structures present in the task provides ways to simplify and speed up reinforcement learning algorithms. These structures also help to generalize such algorithms over multiple tasks without relearning policies from scratch. We use ideas from dynamical systems to find metastable regions in the state space and associate them with abstract states. The spectral clustering algorithm PCCA+ is used to identify suitable abstractions aligned to the underlying structure. Skills are defined in terms of the sequence of actions that lead to transitions between such abstract states. The connectivity information from PCCA+ is used to generate these skills or options. These skills are independent of the learning task and can be efficiently reused across a variety of tasks defined over the same model. This approach works well even without the exact model of the environment by using sample trajectories to construct an approximate estimate. We also present our approach to scaling the skill acquisition framework to complex tasks with large state spaces for which we perform state aggregation using the representation learned from an action conditional video prediction network and use the skill acquisition framework on the aggregated state space.</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, 2020-09
issn 2331-8422
language eng
recordid cdi_proquest_journals_2080574930
source Freely Accessible Journals
subjects Algorithms
Clustering
Machine learning
Metastable region
Skills
Task complexity
title Option Discovery in Hierarchical Reinforcement Learning using Spatio-Temporal Clustering
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T08%3A26%3A54IST&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=Option%20Discovery%20in%20Hierarchical%20Reinforcement%20Learning%20using%20Spatio-Temporal%20Clustering&rft.jtitle=arXiv.org&rft.au=Srinivas,%20Aravind&rft.date=2020-09-21&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2080574930%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2080574930&rft_id=info:pmid/&rfr_iscdi=true