Optimizing Algorithms from Pairwise User Preferences
Typical black-box optimization approaches in robotics focus on learning from metric scores. However, that is not always possible, as not all developers have ground truth available. Learning appropriate robot behavior in human-centric contexts often requires querying users, who typically cannot provi...
Uloženo v:
| Vydáno v: | Proceedings of the ... IEEE/RSJ International Conference on Intelligent Robots and Systems s. 4161 - 4167 |
|---|---|
| Hlavní autoři: | , , , |
| Médium: | Konferenční příspěvek |
| Jazyk: | angličtina |
| Vydáno: |
IEEE
01.10.2023
|
| Témata: | |
| ISSN: | 2153-0866 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | Typical black-box optimization approaches in robotics focus on learning from metric scores. However, that is not always possible, as not all developers have ground truth available. Learning appropriate robot behavior in human-centric contexts often requires querying users, who typically cannot provide precise metric scores. Existing approaches leverage human feedback in an attempt to model an implicit reward function; however, this reward may be difficult or impossible to effectively capture. In this work, we introduce SortCMA to optimize algorithm parameter configurations in high dimensions based on pairwise user preferences. SortCMA efficiently and robustly leverages user input to find parameter sets without directly modeling a reward. We apply this method to tuning a commercial depth sensor without ground truth, and to robot social navigation, which involves highly complex preferences over robot behavior. We show that our method succeeds in optimizing for the user's goals and perform a user study to evaluate social navigation results. |
|---|---|
| AbstractList | Typical black-box optimization approaches in robotics focus on learning from metric scores. However, that is not always possible, as not all developers have ground truth available. Learning appropriate robot behavior in human-centric contexts often requires querying users, who typically cannot provide precise metric scores. Existing approaches leverage human feedback in an attempt to model an implicit reward function; however, this reward may be difficult or impossible to effectively capture. In this work, we introduce SortCMA to optimize algorithm parameter configurations in high dimensions based on pairwise user preferences. SortCMA efficiently and robustly leverages user input to find parameter sets without directly modeling a reward. We apply this method to tuning a commercial depth sensor without ground truth, and to robot social navigation, which involves highly complex preferences over robot behavior. We show that our method succeeds in optimizing for the user's goals and perform a user study to evaluate social navigation results. |
| Author | Shih, Katherine Steinfeld, Aaron Keselman, Leonid Hebert, Martial |
| Author_xml | – sequence: 1 givenname: Leonid surname: Keselman fullname: Keselman, Leonid email: lkeselma@cs.cmu.edu organization: Robotics Institute, School of Computer Science, Carnegie Mellon University,Pittsburgh,PA,USA,15232 – sequence: 2 givenname: Katherine surname: Shih fullname: Shih, Katherine email: kshih@cs.cmu.edu organization: Robotics Institute, School of Computer Science, Carnegie Mellon University,Pittsburgh,PA,USA,15232 – sequence: 3 givenname: Martial surname: Hebert fullname: Hebert, Martial email: hebert@cs.cmu.edu organization: Robotics Institute, School of Computer Science, Carnegie Mellon University,Pittsburgh,PA,USA,15232 – sequence: 4 givenname: Aaron surname: Steinfeld fullname: Steinfeld, Aaron email: astein@cs.cmu.edu organization: Robotics Institute, School of Computer Science, Carnegie Mellon University,Pittsburgh,PA,USA,15232 |
| BookMark | eNo1j91KwzAYQKMouM29gWBfoPP78tfkcgx_BoMOddejSb_MyNqOpCD69ArquTl3B86UXfRDT4zdIiwQwd6tn-sX9QNfcOBigSAkB4NnbIpaK2nRgj5nE45KlGC0vmLznN8BAKGyxuoJk_VpjF38iv2hWB4PQ4rjW5eLkIau2DYxfcRMxS5TKraJAiXqPeVrdhmaY6b5n2ds93D_unoqN_XjerXclJGDHEvuTTDOG29MqKwHQuVb1ZBzrayAo_aOyGrnNOoqcKlIVq6FIHkQ0gopZuzmtxuJaH9KsWvS5_7_UnwDtMNJFA |
| ContentType | Conference Proceeding |
| DBID | 6IE 6IH CBEJK RIE RIO |
| DOI | 10.1109/IROS55552.2023.10342081 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan (POP) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE Electronic Library (IEL) IEEE Proceedings Order Plans (POP) 1998-present |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISBN | 1665491906 9781665491907 |
| EISSN | 2153-0866 |
| EndPage | 4167 |
| ExternalDocumentID | 10342081 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: Office of Naval Research grantid: 90DPGE0003 funderid: 10.13039/100000006 – fundername: National Institute on Disability, Independent Living, and Rehabilitation Research grantid: 90DPGE0003 funderid: 10.13039/100006663 |
| GroupedDBID | 6IE 6IF 6IH 6IL 6IN AAJGR AAWTH ABLEC ACGFS ADZIZ ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ CBEJK CHZPO IEGSK IJVOP M43 OCL RIE RIL RIO RNS |
| ID | FETCH-LOGICAL-i204t-2c8f8bc8c88f79c0e15cd5aebbd470216cbee96bb6167f245e47bd0f42f349343 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 1 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001133658803022&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| IngestDate | Wed Aug 27 02:25:30 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-i204t-2c8f8bc8c88f79c0e15cd5aebbd470216cbee96bb6167f245e47bd0f42f349343 |
| PageCount | 7 |
| ParticipantIDs | ieee_primary_10342081 |
| PublicationCentury | 2000 |
| PublicationDate | 2023-Oct.-1 |
| PublicationDateYYYYMMDD | 2023-10-01 |
| PublicationDate_xml | – month: 10 year: 2023 text: 2023-Oct.-1 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | Proceedings of the ... IEEE/RSJ International Conference on Intelligent Robots and Systems |
| PublicationTitleAbbrev | IROS |
| PublicationYear | 2023 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| SSID | ssj0001079896 |
| Score | 2.2741547 |
| Snippet | Typical black-box optimization approaches in robotics focus on learning from metric scores. However, that is not always possible, as not all developers have... |
| SourceID | ieee |
| SourceType | Publisher |
| StartPage | 4161 |
| SubjectTerms | Behavioral sciences Closed box Measurement Navigation Robot sensing systems Robustness Tuning |
| Title | Optimizing Algorithms from Pairwise User Preferences |
| URI | https://ieeexplore.ieee.org/document/10342081 |
| WOSCitedRecordID | wos001133658803022&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB60eNCLr4pv9uA1NZvNbpKjiEVB2qJWeitJdqIL9kG3VfDXm2xbqwcP5pQEEpJJhmEe3wzAhXSpk4oi0Uoh4coxomysiaCKGu1FuKhKsjzfi1ZL9nqqswCrV1gYRKyCz7ARupUvPx_ZWTCVeQ5PgjfYKzvrQmRzsNbKoEKFkipbxHDFVF3ePbQfU98C3ooljeXqX3VUKjHS3P7nAXagvgLkRZ1vUbMLazjcg60fuQT3gbc98w-KTz-Irt5eRl7pfx2UUYCPRB1dTD6KEqOu_3B-o2Vy2bIO3ebN0_UtWZREIAWjfEqYlU4aK62UTihLMU5tnmo0JufCi-vMGkSVGZPFmXCMp8iFyanjzCVcJTw5gNpwNMRDiCxa5_JUakElN3FIEoNWB49LLHJm8AjqgQD98TzrRX959-M_5k9gM5B5Huh2CrXpZIZnsGHfp0U5Oa_e6gunTJTe |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEG4MmqgXXxjf7sFrsbvbbtujMRKICETBcCNtd4qbCBgWNPHX2y4gevBgT9MmbdppJ5PpzDeD0JWwzApJACspAVNpIyxNqDAnkmjlVDgvSrI8N3izKXo92V6A1QssDAAUwWdQ8WThy0_HZua_ypyEx94b7IyddUYdNYdrrb5UCJdCJosorpDI6_pj64m55hFXUVxZzv9VSaVQJNWdf25hF5VXkLyg_a1s9tAajPbR9o9sggeItpz4D7NP1wluXgdjZ_a_DPPAA0iCtsomH1kOQdc9ObfQMr1sXkbd6l3ntoYXRRFwFhE6xZERVmgjjBCWS0MgZCZlCrROKXcKOzEaQCZaJ2HCbUQZUK5TYmlkYypjGh-i0mg8giMUGDDWpkwoTgTVoU8TA0Z5n0vI00jDMSp7BvTf5nkv-suzn_wxfok2a52HRr9Rb96foi3P8nnY2xkqTSczOEcb5n2a5ZOL4t6-AJfMmCU |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=proceeding&rft.title=Proceedings+of+the+...+IEEE%2FRSJ+International+Conference+on+Intelligent+Robots+and+Systems&rft.atitle=Optimizing+Algorithms+from+Pairwise+User+Preferences&rft.au=Keselman%2C+Leonid&rft.au=Shih%2C+Katherine&rft.au=Hebert%2C+Martial&rft.au=Steinfeld%2C+Aaron&rft.date=2023-10-01&rft.pub=IEEE&rft.eissn=2153-0866&rft.spage=4161&rft.epage=4167&rft_id=info:doi/10.1109%2FIROS55552.2023.10342081&rft.externalDocID=10342081 |