Symbol Grounding and Task Learning from Imperfect Corrections

Mattias Appelgren; Alex Lascarides

doi:10.18653/v1/2021.splurobonlp-1.1

Symbol Grounding and Task Learning from Imperfect Corrections

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract / Description of output

This paper describes a method for learning from a teacher’s potentially unreliable corrective feedback in an interactive task learning setting. The graphical model uses discourse coherence to jointly learn symbol grounding, domain concepts and valid plans. Our experiments show that the agent learns its domain-level task in spite of the teacher’s mistakes.

Original language	English
Title of host publication	Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics
Place of Publication	ACL-IJNLP, 2021
Publisher	Association for Computational Linguistics (ACL)
Pages	1-10
Number of pages	10
ISBN (Electronic)	978-1-954085-78-7
DOIs	https://doi.org/10.18653/v1/2021.splurobonlp-1.1
Publication status	Published - 1 Aug 2021
Event	Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics - Online @ ACL-IJCNLP 2021 Duration: 5 Aug 2021 → 6 Aug 2021 https://splu-robonlp2021.github.io/

Workshop

Workshop	Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics
Abbreviated title	SpLU-RoboNLP 2021
Period	5/08/21 → 6/08/21
Internet address	https://splu-robonlp2021.github.io/

Access to Document

10.18653/v1/2021.splurobonlp-1.1Licence: All Rights Reserved

Symbol Grounding_APPELGREN_DOA29052021_AFVAccepted author manuscript, 662 KBLicence: Creative Commons: Attribution (CC-BY)
Symbol Grounding_APPELGREN_DOA29052021_VOR_CC-BYFinal published version, 791 KBLicence: Creative Commons: Attribution (CC-BY)

https://aclanthology.org/2021.splurobonlp-1.1

Trustworthy Autonomous Systems Regulation and Governance Node
Urquhart, L.
1/11/20 → 30/11/24
Project: Research

Cite this

@inproceedings{fbd80fd786c04877852caf2685414af0,

title = "Symbol Grounding and Task Learning from Imperfect Corrections",

abstract = "This paper describes a method for learning from a teacher{\textquoteright}s potentially unreliable corrective feedback in an interactive task learning setting. The graphical model uses discourse coherence to jointly learn symbol grounding, domain concepts and valid plans. Our experiments show that the agent learns its domain-level task in spite of the teacher{\textquoteright}s mistakes.",

author = "Mattias Appelgren and Alex Lascarides",

year = "2021",

month = aug,

day = "1",

doi = "10.18653/v1/2021.splurobonlp-1.1",

language = "English",

pages = "1--10",

booktitle = "Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics",

publisher = "Association for Computational Linguistics (ACL)",

note = "Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics, SpLU-RoboNLP 2021 ; Conference date: 05-08-2021 Through 06-08-2021",

url = "https://splu-robonlp2021.github.io/",

}

Appelgren, M & Lascarides, A 2021, Symbol Grounding and Task Learning from Imperfect Corrections. in Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics. Association for Computational Linguistics (ACL), ACL-IJNLP, 2021, pp. 1-10, Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics, 5/08/21. https://doi.org/10.18653/v1/2021.splurobonlp-1.1

Symbol Grounding and Task Learning from Imperfect Corrections. / Appelgren, Mattias; Lascarides, Alex.
Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics. ACL-IJNLP, 2021: Association for Computational Linguistics (ACL), 2021. p. 1-10.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Symbol Grounding and Task Learning from Imperfect Corrections

AU - Appelgren, Mattias

AU - Lascarides, Alex

PY - 2021/8/1

Y1 - 2021/8/1

N2 - This paper describes a method for learning from a teacher’s potentially unreliable corrective feedback in an interactive task learning setting. The graphical model uses discourse coherence to jointly learn symbol grounding, domain concepts and valid plans. Our experiments show that the agent learns its domain-level task in spite of the teacher’s mistakes.

AB - This paper describes a method for learning from a teacher’s potentially unreliable corrective feedback in an interactive task learning setting. The graphical model uses discourse coherence to jointly learn symbol grounding, domain concepts and valid plans. Our experiments show that the agent learns its domain-level task in spite of the teacher’s mistakes.

U2 - 10.18653/v1/2021.splurobonlp-1.1

DO - 10.18653/v1/2021.splurobonlp-1.1

M3 - Conference contribution

SP - 1

EP - 10

BT - Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics

PB - Association for Computational Linguistics (ACL)

CY - ACL-IJNLP, 2021

T2 - Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics

Y2 - 5 August 2021 through 6 August 2021

ER -

Symbol Grounding and Task Learning from Imperfect Corrections

Abstract / Description of output

Workshop

Access to Document

Fingerprint

Projects

Trustworthy Autonomous Systems Regulation and Governance Node

Cite this