Image-Based Deep Reinforcement Learning with Intrinsically Motivated Stimuli: On the Execution of Complex Robotic Tasks
Reinforcement Learning (RL) has been widely used to solve tasks where the environment consistently provides a dense reward value. However, in real-world scenarios, rewards can often be poorly defined or sparse. Auxiliary signals are indispensable for discovering efficient exploration strategies and...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2024-07 |
---|---|
Hauptverfasser: | , , , , , |
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 | Valencia, David Williams, Henry Xing, Yuning Gee, Trevor Liarokapis, Minas MacDonald, Bruce A |
description | Reinforcement Learning (RL) has been widely used to solve tasks where the environment consistently provides a dense reward value. However, in real-world scenarios, rewards can often be poorly defined or sparse. Auxiliary signals are indispensable for discovering efficient exploration strategies and aiding the learning process. In this work, inspired by intrinsic motivation theory, we postulate that the intrinsic stimuli of novelty and surprise can assist in improving exploration in complex, sparsely rewarded environments. We introduce a novel sample-efficient method able to learn directly from pixels, an image-based extension of TD3 with an autoencoder called \textit{NaSA-TD3}. The experiments demonstrate that NaSA-TD3 is easy to train and an efficient method for tackling complex continuous-control robotic tasks, both in simulated environments and real-world settings. NaSA-TD3 outperforms existing state-of-the-art RL image-based methods in terms of final performance without requiring pre-trained models or human demonstrations. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_3087032725</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3087032725</sourcerecordid><originalsourceid>FETCH-proquest_journals_30870327253</originalsourceid><addsrcrecordid>eNqNjs1qwkAURodCQWnzDhe6Dkxn_KNLbUoDLYK6lzHe6NXJnXTmptq3bxY-gKtvcc6B70ENjbWv-WxkzEBlKZ201mYyNeOxHapL2bgD5nOXcA_viC2skLgOscIGWeALXWTiA1xIjlCyROJElfP-D76D0K-TPlwLNZ2nN1gyyBGhuGLVCQWGUMMiNK3HK6zCrg8q2Lh0Ts_qsXY-YXbbJ_XyUWwWn3kbw0-HSban0EXu0dbq2VRb0x-291n_6-VNGg</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3087032725</pqid></control><display><type>article</type><title>Image-Based Deep Reinforcement Learning with Intrinsically Motivated Stimuli: On the Execution of Complex Robotic Tasks</title><source>Free E- Journals</source><creator>Valencia, David ; Williams, Henry ; Xing, Yuning ; Gee, Trevor ; Liarokapis, Minas ; MacDonald, Bruce A</creator><creatorcontrib>Valencia, David ; Williams, Henry ; Xing, Yuning ; Gee, Trevor ; Liarokapis, Minas ; MacDonald, Bruce A</creatorcontrib><description>Reinforcement Learning (RL) has been widely used to solve tasks where the environment consistently provides a dense reward value. However, in real-world scenarios, rewards can often be poorly defined or sparse. Auxiliary signals are indispensable for discovering efficient exploration strategies and aiding the learning process. In this work, inspired by intrinsic motivation theory, we postulate that the intrinsic stimuli of novelty and surprise can assist in improving exploration in complex, sparsely rewarded environments. We introduce a novel sample-efficient method able to learn directly from pixels, an image-based extension of TD3 with an autoencoder called \textit{NaSA-TD3}. The experiments demonstrate that NaSA-TD3 is easy to train and an efficient method for tackling complex continuous-control robotic tasks, both in simulated environments and real-world settings. NaSA-TD3 outperforms existing state-of-the-art RL image-based methods in terms of final performance without requiring pre-trained models or human demonstrations.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Deep learning ; Human performance ; Robot control ; Stimuli ; Task complexity</subject><ispartof>arXiv.org, 2024-07</ispartof><rights>2024. 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>Valencia, David</creatorcontrib><creatorcontrib>Williams, Henry</creatorcontrib><creatorcontrib>Xing, Yuning</creatorcontrib><creatorcontrib>Gee, Trevor</creatorcontrib><creatorcontrib>Liarokapis, Minas</creatorcontrib><creatorcontrib>MacDonald, Bruce A</creatorcontrib><title>Image-Based Deep Reinforcement Learning with Intrinsically Motivated Stimuli: On the Execution of Complex Robotic Tasks</title><title>arXiv.org</title><description>Reinforcement Learning (RL) has been widely used to solve tasks where the environment consistently provides a dense reward value. However, in real-world scenarios, rewards can often be poorly defined or sparse. Auxiliary signals are indispensable for discovering efficient exploration strategies and aiding the learning process. In this work, inspired by intrinsic motivation theory, we postulate that the intrinsic stimuli of novelty and surprise can assist in improving exploration in complex, sparsely rewarded environments. We introduce a novel sample-efficient method able to learn directly from pixels, an image-based extension of TD3 with an autoencoder called \textit{NaSA-TD3}. The experiments demonstrate that NaSA-TD3 is easy to train and an efficient method for tackling complex continuous-control robotic tasks, both in simulated environments and real-world settings. NaSA-TD3 outperforms existing state-of-the-art RL image-based methods in terms of final performance without requiring pre-trained models or human demonstrations.</description><subject>Deep learning</subject><subject>Human performance</subject><subject>Robot control</subject><subject>Stimuli</subject><subject>Task complexity</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNjs1qwkAURodCQWnzDhe6Dkxn_KNLbUoDLYK6lzHe6NXJnXTmptq3bxY-gKtvcc6B70ENjbWv-WxkzEBlKZ201mYyNeOxHapL2bgD5nOXcA_viC2skLgOscIGWeALXWTiA1xIjlCyROJElfP-D76D0K-TPlwLNZ2nN1gyyBGhuGLVCQWGUMMiNK3HK6zCrg8q2Lh0Ts_qsXY-YXbbJ_XyUWwWn3kbw0-HSban0EXu0dbq2VRb0x-291n_6-VNGg</recordid><startdate>20240731</startdate><enddate>20240731</enddate><creator>Valencia, David</creator><creator>Williams, Henry</creator><creator>Xing, Yuning</creator><creator>Gee, Trevor</creator><creator>Liarokapis, Minas</creator><creator>MacDonald, Bruce A</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>20240731</creationdate><title>Image-Based Deep Reinforcement Learning with Intrinsically Motivated Stimuli: On the Execution of Complex Robotic Tasks</title><author>Valencia, David ; Williams, Henry ; Xing, Yuning ; Gee, Trevor ; Liarokapis, Minas ; MacDonald, Bruce A</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_30870327253</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Deep learning</topic><topic>Human performance</topic><topic>Robot control</topic><topic>Stimuli</topic><topic>Task complexity</topic><toplevel>online_resources</toplevel><creatorcontrib>Valencia, David</creatorcontrib><creatorcontrib>Williams, Henry</creatorcontrib><creatorcontrib>Xing, Yuning</creatorcontrib><creatorcontrib>Gee, Trevor</creatorcontrib><creatorcontrib>Liarokapis, Minas</creatorcontrib><creatorcontrib>MacDonald, Bruce A</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & 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>Valencia, David</au><au>Williams, Henry</au><au>Xing, Yuning</au><au>Gee, Trevor</au><au>Liarokapis, Minas</au><au>MacDonald, Bruce A</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Image-Based Deep Reinforcement Learning with Intrinsically Motivated Stimuli: On the Execution of Complex Robotic Tasks</atitle><jtitle>arXiv.org</jtitle><date>2024-07-31</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Reinforcement Learning (RL) has been widely used to solve tasks where the environment consistently provides a dense reward value. However, in real-world scenarios, rewards can often be poorly defined or sparse. Auxiliary signals are indispensable for discovering efficient exploration strategies and aiding the learning process. In this work, inspired by intrinsic motivation theory, we postulate that the intrinsic stimuli of novelty and surprise can assist in improving exploration in complex, sparsely rewarded environments. We introduce a novel sample-efficient method able to learn directly from pixels, an image-based extension of TD3 with an autoencoder called \textit{NaSA-TD3}. The experiments demonstrate that NaSA-TD3 is easy to train and an efficient method for tackling complex continuous-control robotic tasks, both in simulated environments and real-world settings. NaSA-TD3 outperforms existing state-of-the-art RL image-based methods in terms of final performance without requiring pre-trained models or human demonstrations.</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, 2024-07 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_3087032725 |
source | Free E- Journals |
subjects | Deep learning Human performance Robot control Stimuli Task complexity |
title | Image-Based Deep Reinforcement Learning with Intrinsically Motivated Stimuli: On the Execution of Complex Robotic Tasks |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-02T23%3A20%3A33IST&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=Image-Based%20Deep%20Reinforcement%20Learning%20with%20Intrinsically%20Motivated%20Stimuli:%20On%20the%20Execution%20of%20Complex%20Robotic%20Tasks&rft.jtitle=arXiv.org&rft.au=Valencia,%20David&rft.date=2024-07-31&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E3087032725%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3087032725&rft_id=info:pmid/&rfr_iscdi=true |