Active Curriculum Learning

Borna Jafarpour; Nicolai Pogrebnyakov; Dawn  Sepehr

Active Curriculum Learning

Borna Jafarpour^*, Nicolai Pogrebnyakov, Dawn Sepehr

^*Corresponding author af dette arbejde

Institut for International Økonomi, Politik og Business

Thomson Reuters Canada

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › peer review

Abstract

This paper investigates and reveals the relationship between two closely related machine learning disciplines, namely Active Learning (AL) and Curriculum Learning (CL), from the lens of several novel curricula. This paper also introduces Active Curriculum Learning (ACL) which improves AL by combining AL with CL to benefit from the dynamic nature of the AL informativeness concept as well as the human insights used in the design of the curriculum heuristics. Comparison of the performance of ACL and AL on two public datasets for the Named Entity Recognition (NER) task shows the effectiveness of combining AL and CL using our proposed framework.

Originalsprog	Engelsk
Titel	InterNLP 2021. First Workshop on Interactive Learning for Natural Language Processing : Proceedings of the Workshop
Antal sider	6
Udgivelsessted	Stroudsburg, PA
Forlag	Association for Computational Linguistics
Publikationsdato	2021
Sider	40-45
ISBN (Trykt)	9781954085664
Status	Udgivet - 2021
Begivenhed	First Workshop on Interactive Learning for Natural Language Processing. InterNLP 2021: Workshop at ACL 2021 - Virtual Conference Varighed: 5 aug. 2021 → 5 aug. 2021 Konferencens nummer: 1 https://sites.google.com/view/internlp2021/home

Workshop

Workshop	First Workshop on Interactive Learning for Natural Language Processing. InterNLP 2021
Nummer	1
Lokation	Virtual Conference
Periode	05/08/2021 → 05/08/2021
Internetadresse	https://sites.google.com/view/internlp2021/home

Adgang til dokumentet

https://aclanthology.org/2021.internlp-1.pdfLicens: Ikke-specificeret

Handle.net

Persistent link

Citationsformater

@inproceedings{235a7905740e40f2879d3f6ae085e8d3,

title = "Active Curriculum Learning",

abstract = "This paper investigates and reveals the relationship between two closely related machine learning disciplines, namely Active Learning (AL) and Curriculum Learning (CL), from the lens of several novel curricula. This paper also introduces Active Curriculum Learning (ACL) which improves AL by combining AL with CL to benefit from the dynamic nature of the AL informativeness concept as well as the human insights used in the design of the curriculum heuristics. Comparison of the performance of ACL and AL on two public datasets for the Named Entity Recognition (NER) task shows the effectiveness of combining AL and CL using our proposed framework.",

author = "Borna Jafarpour and Nicolai Pogrebnyakov and Dawn Sepehr",

year = "2021",

language = "English",

isbn = "9781954085664",

pages = "40--45",

booktitle = "InterNLP 2021. First Workshop on Interactive Learning for Natural Language Processing",

publisher = "Association for Computational Linguistics",

address = "United States",

note = "First Workshop on Interactive Learning for Natural Language Processing. InterNLP 2021 ; Conference date: 05-08-2021 Through 05-08-2021",

url = "https://sites.google.com/view/internlp2021/home",

}

Jafarpour, B, Pogrebnyakov, N & Sepehr, D 2021, Active Curriculum Learning. i InterNLP 2021. First Workshop on Interactive Learning for Natural Language Processing: Proceedings of the Workshop. Association for Computational Linguistics, Stroudsburg, PA , s. 40-45, First Workshop on Interactive Learning for Natural Language Processing. InterNLP 2021, 05/08/2021. <https://aclanthology.org/2021.internlp-1.pdf>

Active Curriculum Learning. / Jafarpour, Borna; Pogrebnyakov, Nicolai; Sepehr, Dawn .
InterNLP 2021. First Workshop on Interactive Learning for Natural Language Processing: Proceedings of the Workshop. Stroudsburg, PA : Association for Computational Linguistics, 2021. s. 40-45.

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › peer review

TY - GEN

T1 - Active Curriculum Learning

AU - Jafarpour, Borna

AU - Pogrebnyakov, Nicolai

AU - Sepehr, Dawn

N1 - Conference code: 1

PY - 2021

Y1 - 2021

N2 - This paper investigates and reveals the relationship between two closely related machine learning disciplines, namely Active Learning (AL) and Curriculum Learning (CL), from the lens of several novel curricula. This paper also introduces Active Curriculum Learning (ACL) which improves AL by combining AL with CL to benefit from the dynamic nature of the AL informativeness concept as well as the human insights used in the design of the curriculum heuristics. Comparison of the performance of ACL and AL on two public datasets for the Named Entity Recognition (NER) task shows the effectiveness of combining AL and CL using our proposed framework.

AB - This paper investigates and reveals the relationship between two closely related machine learning disciplines, namely Active Learning (AL) and Curriculum Learning (CL), from the lens of several novel curricula. This paper also introduces Active Curriculum Learning (ACL) which improves AL by combining AL with CL to benefit from the dynamic nature of the AL informativeness concept as well as the human insights used in the design of the curriculum heuristics. Comparison of the performance of ACL and AL on two public datasets for the Named Entity Recognition (NER) task shows the effectiveness of combining AL and CL using our proposed framework.

M3 - Article in proceedings

SN - 9781954085664

SP - 40

EP - 45

BT - InterNLP 2021. First Workshop on Interactive Learning for Natural Language Processing

PB - Association for Computational Linguistics

CY - Stroudsburg, PA

T2 - First Workshop on Interactive Learning for Natural Language Processing. InterNLP 2021

Y2 - 5 August 2021 through 5 August 2021

ER -