Grid Cell-Inspired Fragmentation and Recall for Efficient Map Building

Animals and robots navigate through environments by building and refining maps of space. These maps enable functions including navigation back to home, planning, search and foraging. Here, we use observations from neuroscience, specifically the observed fragmentation of grid cell map in compartmenta...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hwang, Jaedong, Hong, Zhang-Wei, Chen, Eric, Boopathy, Akhilan, Agrawal, Pulkit, Fiete, Ila
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Hwang, Jaedong
Hong, Zhang-Wei
Chen, Eric
Boopathy, Akhilan
Agrawal, Pulkit
Fiete, Ila
description Animals and robots navigate through environments by building and refining maps of space. These maps enable functions including navigation back to home, planning, search and foraging. Here, we use observations from neuroscience, specifically the observed fragmentation of grid cell map in compartmentalized spaces, to propose and apply the concept of Fragmentation-and-Recall (FARMap) in the mapping of large spaces. Agents solve the mapping problem by building local maps via a surprisal-based clustering of space, which they use to set subgoals for spatial exploration. Agents build and use a local map to predict their observations; high surprisal leads to a "fragmentation event" that truncates the local map. At these events, the recent local map is placed into long-term memory (LTM) and a different local map is initialized. If observations at a fracture point match observations in one of the stored local maps, that map is recalled (and thus reused) from LTM. The fragmentation points induce a natural online clustering of the larger space, forming a set of intrinsic potential subgoals that are stored in LTM as a topological graph. Agents choose their next subgoal from the set of near and far potential subgoals from within the current local map or LTM, respectively. Thus, local maps guide exploration locally, while LTM promotes global exploration. We demonstrate that FARMap replicates the fragmentation points observed in animal studies. We evaluate FARMap on complex procedurally-generated spatial environments and realistic simulations to demonstrate that this mapping strategy much more rapidly covers the environment (number of agent steps and wall clock time) and is more efficient in active memory usage, without loss of performance. https://jd730.github.io/projects/FARMap/
doi_str_mv 10.48550/arxiv.2307.05793
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2307_05793</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2307_05793</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-6c00f8757e915b71c3d7b1d18475c9de3dddce03bf61460a54be5f30d7abb8f23</originalsourceid><addsrcrecordid>eNotz8FKAzEUheFsXEj1AVyZF5jxpplMZpY6dGqhIkj3w03uTQmk6ZBW0bdXq6uz-OHAJ8SdgrrpjIEHLJ_xo15qsDUY2-trMa5LJDlwStUmn-ZYmORYcH_gfMZzPGaJmeQbe0xJhmORqxCijz9VvuAsn95jopj3N-IqYDrx7f8uxG5c7Ybnavu63gyP2wpbq6vWA4TOGsu9Ms4qr8k6RaprrPE9sSYiz6BdaFXTAprGsQkayKJzXVjqhbj_u71AprnEA5av6Rc0XUD6G5XgRcg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Grid Cell-Inspired Fragmentation and Recall for Efficient Map Building</title><source>arXiv.org</source><creator>Hwang, Jaedong ; Hong, Zhang-Wei ; Chen, Eric ; Boopathy, Akhilan ; Agrawal, Pulkit ; Fiete, Ila</creator><creatorcontrib>Hwang, Jaedong ; Hong, Zhang-Wei ; Chen, Eric ; Boopathy, Akhilan ; Agrawal, Pulkit ; Fiete, Ila</creatorcontrib><description>Animals and robots navigate through environments by building and refining maps of space. These maps enable functions including navigation back to home, planning, search and foraging. Here, we use observations from neuroscience, specifically the observed fragmentation of grid cell map in compartmentalized spaces, to propose and apply the concept of Fragmentation-and-Recall (FARMap) in the mapping of large spaces. Agents solve the mapping problem by building local maps via a surprisal-based clustering of space, which they use to set subgoals for spatial exploration. Agents build and use a local map to predict their observations; high surprisal leads to a "fragmentation event" that truncates the local map. At these events, the recent local map is placed into long-term memory (LTM) and a different local map is initialized. If observations at a fracture point match observations in one of the stored local maps, that map is recalled (and thus reused) from LTM. The fragmentation points induce a natural online clustering of the larger space, forming a set of intrinsic potential subgoals that are stored in LTM as a topological graph. Agents choose their next subgoal from the set of near and far potential subgoals from within the current local map or LTM, respectively. Thus, local maps guide exploration locally, while LTM promotes global exploration. We demonstrate that FARMap replicates the fragmentation points observed in animal studies. We evaluate FARMap on complex procedurally-generated spatial environments and realistic simulations to demonstrate that this mapping strategy much more rapidly covers the environment (number of agent steps and wall clock time) and is more efficient in active memory usage, without loss of performance. https://jd730.github.io/projects/FARMap/</description><identifier>DOI: 10.48550/arxiv.2307.05793</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Robotics</subject><creationdate>2023-07</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2307.05793$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2307.05793$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Hwang, Jaedong</creatorcontrib><creatorcontrib>Hong, Zhang-Wei</creatorcontrib><creatorcontrib>Chen, Eric</creatorcontrib><creatorcontrib>Boopathy, Akhilan</creatorcontrib><creatorcontrib>Agrawal, Pulkit</creatorcontrib><creatorcontrib>Fiete, Ila</creatorcontrib><title>Grid Cell-Inspired Fragmentation and Recall for Efficient Map Building</title><description>Animals and robots navigate through environments by building and refining maps of space. These maps enable functions including navigation back to home, planning, search and foraging. Here, we use observations from neuroscience, specifically the observed fragmentation of grid cell map in compartmentalized spaces, to propose and apply the concept of Fragmentation-and-Recall (FARMap) in the mapping of large spaces. Agents solve the mapping problem by building local maps via a surprisal-based clustering of space, which they use to set subgoals for spatial exploration. Agents build and use a local map to predict their observations; high surprisal leads to a "fragmentation event" that truncates the local map. At these events, the recent local map is placed into long-term memory (LTM) and a different local map is initialized. If observations at a fracture point match observations in one of the stored local maps, that map is recalled (and thus reused) from LTM. The fragmentation points induce a natural online clustering of the larger space, forming a set of intrinsic potential subgoals that are stored in LTM as a topological graph. Agents choose their next subgoal from the set of near and far potential subgoals from within the current local map or LTM, respectively. Thus, local maps guide exploration locally, while LTM promotes global exploration. We demonstrate that FARMap replicates the fragmentation points observed in animal studies. We evaluate FARMap on complex procedurally-generated spatial environments and realistic simulations to demonstrate that this mapping strategy much more rapidly covers the environment (number of agent steps and wall clock time) and is more efficient in active memory usage, without loss of performance. https://jd730.github.io/projects/FARMap/</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Robotics</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz8FKAzEUheFsXEj1AVyZF5jxpplMZpY6dGqhIkj3w03uTQmk6ZBW0bdXq6uz-OHAJ8SdgrrpjIEHLJ_xo15qsDUY2-trMa5LJDlwStUmn-ZYmORYcH_gfMZzPGaJmeQbe0xJhmORqxCijz9VvuAsn95jopj3N-IqYDrx7f8uxG5c7Ybnavu63gyP2wpbq6vWA4TOGsu9Ms4qr8k6RaprrPE9sSYiz6BdaFXTAprGsQkayKJzXVjqhbj_u71AprnEA5av6Rc0XUD6G5XgRcg</recordid><startdate>20230711</startdate><enddate>20230711</enddate><creator>Hwang, Jaedong</creator><creator>Hong, Zhang-Wei</creator><creator>Chen, Eric</creator><creator>Boopathy, Akhilan</creator><creator>Agrawal, Pulkit</creator><creator>Fiete, Ila</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230711</creationdate><title>Grid Cell-Inspired Fragmentation and Recall for Efficient Map Building</title><author>Hwang, Jaedong ; Hong, Zhang-Wei ; Chen, Eric ; Boopathy, Akhilan ; Agrawal, Pulkit ; Fiete, Ila</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-6c00f8757e915b71c3d7b1d18475c9de3dddce03bf61460a54be5f30d7abb8f23</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Robotics</topic><toplevel>online_resources</toplevel><creatorcontrib>Hwang, Jaedong</creatorcontrib><creatorcontrib>Hong, Zhang-Wei</creatorcontrib><creatorcontrib>Chen, Eric</creatorcontrib><creatorcontrib>Boopathy, Akhilan</creatorcontrib><creatorcontrib>Agrawal, Pulkit</creatorcontrib><creatorcontrib>Fiete, Ila</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Hwang, Jaedong</au><au>Hong, Zhang-Wei</au><au>Chen, Eric</au><au>Boopathy, Akhilan</au><au>Agrawal, Pulkit</au><au>Fiete, Ila</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Grid Cell-Inspired Fragmentation and Recall for Efficient Map Building</atitle><date>2023-07-11</date><risdate>2023</risdate><abstract>Animals and robots navigate through environments by building and refining maps of space. These maps enable functions including navigation back to home, planning, search and foraging. Here, we use observations from neuroscience, specifically the observed fragmentation of grid cell map in compartmentalized spaces, to propose and apply the concept of Fragmentation-and-Recall (FARMap) in the mapping of large spaces. Agents solve the mapping problem by building local maps via a surprisal-based clustering of space, which they use to set subgoals for spatial exploration. Agents build and use a local map to predict their observations; high surprisal leads to a "fragmentation event" that truncates the local map. At these events, the recent local map is placed into long-term memory (LTM) and a different local map is initialized. If observations at a fracture point match observations in one of the stored local maps, that map is recalled (and thus reused) from LTM. The fragmentation points induce a natural online clustering of the larger space, forming a set of intrinsic potential subgoals that are stored in LTM as a topological graph. Agents choose their next subgoal from the set of near and far potential subgoals from within the current local map or LTM, respectively. Thus, local maps guide exploration locally, while LTM promotes global exploration. We demonstrate that FARMap replicates the fragmentation points observed in animal studies. We evaluate FARMap on complex procedurally-generated spatial environments and realistic simulations to demonstrate that this mapping strategy much more rapidly covers the environment (number of agent steps and wall clock time) and is more efficient in active memory usage, without loss of performance. https://jd730.github.io/projects/FARMap/</abstract><doi>10.48550/arxiv.2307.05793</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2307.05793
ispartof
issn
language eng
recordid cdi_arxiv_primary_2307_05793
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Robotics
title Grid Cell-Inspired Fragmentation and Recall for Efficient Map Building
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-28T10%3A16%3A26IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Grid%20Cell-Inspired%20Fragmentation%20and%20Recall%20for%20Efficient%20Map%20Building&rft.au=Hwang,%20Jaedong&rft.date=2023-07-11&rft_id=info:doi/10.48550/arxiv.2307.05793&rft_dat=%3Carxiv_GOX%3E2307_05793%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true