Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities

Contemporary artificial intelligence systems exhibit rapidly growing abilities accompanied by the growth of required resources, expansive datasets and corresponding investments into computing infrastructure. Although earlier successes predominantly focus on constrained settings, recent strides in fu...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wulfmeier, Markus, Byravan, Arunkumar, Bechtle, Sarah, Hausman, Karol, Heess, Nicolas
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 Wulfmeier, Markus
Byravan, Arunkumar
Bechtle, Sarah
Hausman, Karol
Heess, Nicolas
description Contemporary artificial intelligence systems exhibit rapidly growing abilities accompanied by the growth of required resources, expansive datasets and corresponding investments into computing infrastructure. Although earlier successes predominantly focus on constrained settings, recent strides in fundamental research and applications aspire to create increasingly general systems. This evolving landscape presents a dual panorama of opportunities and challenges in refining the generalisation and transfer of knowledge - the extraction from existing sources and adaptation as a comprehensive foundation for tackling new problems. Within the domain of reinforcement learning (RL), the representation of knowledge manifests through various modalities, including dynamics and reward models, value functions, policies, and the original data. This taxonomy systematically targets these modalities and frames its discussion based on their inherent properties and alignment with different objectives and mechanisms for transfer. Where possible, we aim to provide coarse guidance delineating approaches which address requirements such as limiting environment interactions, maximising computational efficiency, and enhancing generalisation across varying axes of change. Finally, we analyse reasons contributing to the prevalence or scarcity of specific forms of transfer, the inherent potential behind pushing these frontiers, and underscore the significance of transitioning from designed to learned transfer.
doi_str_mv 10.48550/arxiv.2312.01939
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2312_01939</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2312_01939</sourcerecordid><originalsourceid>FETCH-LOGICAL-a679-79ee9409de9bfeef70352629bc1b0519bfcbac696c029f2bc5b06c8c088f7f993</originalsourceid><addsrcrecordid>eNotz09LwzAcxvFcPMj0BXgyb6A1f5a0v93GcCpWBCm7liT9ZQTaRNKq27t3zp0e-B4e-BByx1m5rJViDyYfwncpJBcl4yDhmuy26Sv2Zg4pTtSnTNts4uQx0xDpB4Z4ag5HjDNt0OQY4n5F17Q1hxTTeKTJ09eYfgbs90jfUm-GMAecbsiVN8OEt5ddkHb72G6ei-b96WWzbgqjKygqQIQlgx7BekRfMamEFmAdt0zxU3TWOA3aMQFeWKcs0652rK595QHkgtz_355h3WcOo8nH7g_YnYHyF6bHTJg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities</title><source>arXiv.org</source><creator>Wulfmeier, Markus ; Byravan, Arunkumar ; Bechtle, Sarah ; Hausman, Karol ; Heess, Nicolas</creator><creatorcontrib>Wulfmeier, Markus ; Byravan, Arunkumar ; Bechtle, Sarah ; Hausman, Karol ; Heess, Nicolas</creatorcontrib><description>Contemporary artificial intelligence systems exhibit rapidly growing abilities accompanied by the growth of required resources, expansive datasets and corresponding investments into computing infrastructure. Although earlier successes predominantly focus on constrained settings, recent strides in fundamental research and applications aspire to create increasingly general systems. This evolving landscape presents a dual panorama of opportunities and challenges in refining the generalisation and transfer of knowledge - the extraction from existing sources and adaptation as a comprehensive foundation for tackling new problems. Within the domain of reinforcement learning (RL), the representation of knowledge manifests through various modalities, including dynamics and reward models, value functions, policies, and the original data. This taxonomy systematically targets these modalities and frames its discussion based on their inherent properties and alignment with different objectives and mechanisms for transfer. Where possible, we aim to provide coarse guidance delineating approaches which address requirements such as limiting environment interactions, maximising computational efficiency, and enhancing generalisation across varying axes of change. Finally, we analyse reasons contributing to the prevalence or scarcity of specific forms of transfer, the inherent potential behind pushing these frontiers, and underscore the significance of transitioning from designed to learned transfer.</description><identifier>DOI: 10.48550/arxiv.2312.01939</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Learning ; Computer Science - Robotics ; Statistics - Machine Learning</subject><creationdate>2023-12</creationdate><rights>http://creativecommons.org/licenses/by-nc-sa/4.0</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/2312.01939$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2312.01939$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Wulfmeier, Markus</creatorcontrib><creatorcontrib>Byravan, Arunkumar</creatorcontrib><creatorcontrib>Bechtle, Sarah</creatorcontrib><creatorcontrib>Hausman, Karol</creatorcontrib><creatorcontrib>Heess, Nicolas</creatorcontrib><title>Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities</title><description>Contemporary artificial intelligence systems exhibit rapidly growing abilities accompanied by the growth of required resources, expansive datasets and corresponding investments into computing infrastructure. Although earlier successes predominantly focus on constrained settings, recent strides in fundamental research and applications aspire to create increasingly general systems. This evolving landscape presents a dual panorama of opportunities and challenges in refining the generalisation and transfer of knowledge - the extraction from existing sources and adaptation as a comprehensive foundation for tackling new problems. Within the domain of reinforcement learning (RL), the representation of knowledge manifests through various modalities, including dynamics and reward models, value functions, policies, and the original data. This taxonomy systematically targets these modalities and frames its discussion based on their inherent properties and alignment with different objectives and mechanisms for transfer. Where possible, we aim to provide coarse guidance delineating approaches which address requirements such as limiting environment interactions, maximising computational efficiency, and enhancing generalisation across varying axes of change. Finally, we analyse reasons contributing to the prevalence or scarcity of specific forms of transfer, the inherent potential behind pushing these frontiers, and underscore the significance of transitioning from designed to learned transfer.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Learning</subject><subject>Computer Science - Robotics</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz09LwzAcxvFcPMj0BXgyb6A1f5a0v93GcCpWBCm7liT9ZQTaRNKq27t3zp0e-B4e-BByx1m5rJViDyYfwncpJBcl4yDhmuy26Sv2Zg4pTtSnTNts4uQx0xDpB4Z4ag5HjDNt0OQY4n5F17Q1hxTTeKTJ09eYfgbs90jfUm-GMAecbsiVN8OEt5ddkHb72G6ei-b96WWzbgqjKygqQIQlgx7BekRfMamEFmAdt0zxU3TWOA3aMQFeWKcs0652rK595QHkgtz_355h3WcOo8nH7g_YnYHyF6bHTJg</recordid><startdate>20231204</startdate><enddate>20231204</enddate><creator>Wulfmeier, Markus</creator><creator>Byravan, Arunkumar</creator><creator>Bechtle, Sarah</creator><creator>Hausman, Karol</creator><creator>Heess, Nicolas</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20231204</creationdate><title>Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities</title><author>Wulfmeier, Markus ; Byravan, Arunkumar ; Bechtle, Sarah ; Hausman, Karol ; Heess, Nicolas</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a679-79ee9409de9bfeef70352629bc1b0519bfcbac696c029f2bc5b06c8c088f7f993</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><topic>Computer Science - Robotics</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Wulfmeier, Markus</creatorcontrib><creatorcontrib>Byravan, Arunkumar</creatorcontrib><creatorcontrib>Bechtle, Sarah</creatorcontrib><creatorcontrib>Hausman, Karol</creatorcontrib><creatorcontrib>Heess, Nicolas</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Wulfmeier, Markus</au><au>Byravan, Arunkumar</au><au>Bechtle, Sarah</au><au>Hausman, Karol</au><au>Heess, Nicolas</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities</atitle><date>2023-12-04</date><risdate>2023</risdate><abstract>Contemporary artificial intelligence systems exhibit rapidly growing abilities accompanied by the growth of required resources, expansive datasets and corresponding investments into computing infrastructure. Although earlier successes predominantly focus on constrained settings, recent strides in fundamental research and applications aspire to create increasingly general systems. This evolving landscape presents a dual panorama of opportunities and challenges in refining the generalisation and transfer of knowledge - the extraction from existing sources and adaptation as a comprehensive foundation for tackling new problems. Within the domain of reinforcement learning (RL), the representation of knowledge manifests through various modalities, including dynamics and reward models, value functions, policies, and the original data. This taxonomy systematically targets these modalities and frames its discussion based on their inherent properties and alignment with different objectives and mechanisms for transfer. Where possible, we aim to provide coarse guidance delineating approaches which address requirements such as limiting environment interactions, maximising computational efficiency, and enhancing generalisation across varying axes of change. Finally, we analyse reasons contributing to the prevalence or scarcity of specific forms of transfer, the inherent potential behind pushing these frontiers, and underscore the significance of transitioning from designed to learned transfer.</abstract><doi>10.48550/arxiv.2312.01939</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2312.01939
ispartof
issn
language eng
recordid cdi_arxiv_primary_2312_01939
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Learning
Computer Science - Robotics
Statistics - Machine Learning
title Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T05%3A12%3A47IST&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=Foundations%20for%20Transfer%20in%20Reinforcement%20Learning:%20A%20Taxonomy%20of%20Knowledge%20Modalities&rft.au=Wulfmeier,%20Markus&rft.date=2023-12-04&rft_id=info:doi/10.48550/arxiv.2312.01939&rft_dat=%3Carxiv_GOX%3E2312_01939%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