Continuation methods and curriculum learning for learning to rank

Nicola Ferro; Maria Maistro; Claudio Lucchese; Raffaele Perego

doi:10.1145/3269206.3269239

Continuation methods and curriculum learning for learning to rank

Nicola Ferro, Maria Maistro, Claudio Lucchese, Raffaele Perego

4 Citations (Scopus)

Abstract

In this paper we explore the use of Continuation Methods and Curriculum Learning techniques in the area of Learning to Rank. The basic idea is to design the training process as a learning path across increasingly complex training instances and objective functions. We propose to instantiate continuation methods in Learning to Rank by changing the IR measure to optimize during training, and we present two different curriculum learning strategies to identify easy training examples. Experimental results show that simple continuation methods are more promising than curriculum learning ones since they allow for slightly improving the performance of state-of-the-art ?-MART models and provide a faster convergence speed.

Original language	English
Title of host publication	CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management
Editors	Norman Paton, Selcuk Candan, Haixun Wang, James Allan, Rakesh Agrawal, Alexandros Labrinidis, Alfredo Cuzzocrea, Mohammed Zaki, Divesh Srivastava, Andrei Broder, Assaf Schuster
Number of pages	4
Publisher	Association for Computing Machinery, Inc.
Publication date	17 Oct 2018
Pages	1523-1526
ISBN (Electronic)	9781450360142
DOIs	https://doi.org/10.1145/3269206.3269239
Publication status	Published - 17 Oct 2018
Externally published	Yes
Event	27th ACM International Conference on Information and Knowledge Management, CIKM 2018 - Torino, Italy Duration: 22 Oct 2018 → 26 Oct 2018

Conference

Conference	27th ACM International Conference on Information and Knowledge Management, CIKM 2018
Country/Territory	Italy
City	Torino
Period	22/10/2018 → 26/10/2018
Sponsor	ACM SIGIR, ACM SIGWEB

Keywords

Curriculum learning
Lambdamart
Learning to rank

Access to Document

10.1145/3269206.3269239

Cite this

Ferro, N., Maistro, M., Lucchese, C., & Perego, R. (2018). Continuation methods and curriculum learning for learning to rank. In N. Paton, S. Candan, H. Wang, J. Allan, R. Agrawal, A. Labrinidis, A. Cuzzocrea, M. Zaki, D. Srivastava, A. Broder, & A. Schuster (Eds.), CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management (pp. 1523-1526). Association for Computing Machinery, Inc.. https://doi.org/10.1145/3269206.3269239

Continuation methods and curriculum learning for learning to rank. / Ferro, Nicola; Maistro, Maria; Lucchese, Claudio et al.

CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. ed. / Norman Paton; Selcuk Candan; Haixun Wang; James Allan; Rakesh Agrawal; Alexandros Labrinidis; Alfredo Cuzzocrea; Mohammed Zaki; Divesh Srivastava; Andrei Broder; Assaf Schuster. Association for Computing Machinery, Inc., 2018. p. 1523-1526.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Ferro, N, Maistro, M, Lucchese, C & Perego, R 2018, Continuation methods and curriculum learning for learning to rank. in N Paton, S Candan, H Wang, J Allan, R Agrawal, A Labrinidis, A Cuzzocrea, M Zaki, D Srivastava, A Broder & A Schuster (eds), CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, Inc., pp. 1523-1526, 27th ACM International Conference on Information and Knowledge Management, CIKM 2018, Torino, Italy, 22/10/2018. https://doi.org/10.1145/3269206.3269239

Ferro N, Maistro M, Lucchese C, Perego R. Continuation methods and curriculum learning for learning to rank. In Paton N, Candan S, Wang H, Allan J, Agrawal R, Labrinidis A, Cuzzocrea A, Zaki M, Srivastava D, Broder A, Schuster A, editors, CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery, Inc. 2018. p. 1523-1526 doi: 10.1145/3269206.3269239

Ferro, Nicola ; Maistro, Maria ; Lucchese, Claudio et al. / Continuation methods and curriculum learning for learning to rank. CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management. editor / Norman Paton ; Selcuk Candan ; Haixun Wang ; James Allan ; Rakesh Agrawal ; Alexandros Labrinidis ; Alfredo Cuzzocrea ; Mohammed Zaki ; Divesh Srivastava ; Andrei Broder ; Assaf Schuster. Association for Computing Machinery, Inc., 2018. pp. 1523-1526

@inproceedings{3d69e6199a5142c3a5d9d4d7093f84c2,

title = "Continuation methods and curriculum learning for learning to rank",

abstract = "In this paper we explore the use of Continuation Methods and Curriculum Learning techniques in the area of Learning to Rank. The basic idea is to design the training process as a learning path across increasingly complex training instances and objective functions. We propose to instantiate continuation methods in Learning to Rank by changing the IR measure to optimize during training, and we present two different curriculum learning strategies to identify easy training examples. Experimental results show that simple continuation methods are more promising than curriculum learning ones since they allow for slightly improving the performance of state-of-the-art ?-MART models and provide a faster convergence speed.",

keywords = "Curriculum learning, Lambdamart, Learning to rank",

author = "Nicola Ferro and Maria Maistro and Claudio Lucchese and Raffaele Perego",

year = "2018",

month = oct,

day = "17",

doi = "10.1145/3269206.3269239",

language = "English",

pages = "1523--1526",

editor = "Norman Paton and Selcuk Candan and Haixun Wang and James Allan and Rakesh Agrawal and Alexandros Labrinidis and Alfredo Cuzzocrea and Mohammed Zaki and Divesh Srivastava and Andrei Broder and Assaf Schuster",

booktitle = "CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management",

publisher = "Association for Computing Machinery, Inc.",

note = "27th ACM International Conference on Information and Knowledge Management, CIKM 2018 ; Conference date: 22-10-2018 Through 26-10-2018",

}

TY - GEN

T1 - Continuation methods and curriculum learning for learning to rank

AU - Ferro, Nicola

AU - Maistro, Maria

AU - Lucchese, Claudio

AU - Perego, Raffaele

PY - 2018/10/17

Y1 - 2018/10/17

N2 - In this paper we explore the use of Continuation Methods and Curriculum Learning techniques in the area of Learning to Rank. The basic idea is to design the training process as a learning path across increasingly complex training instances and objective functions. We propose to instantiate continuation methods in Learning to Rank by changing the IR measure to optimize during training, and we present two different curriculum learning strategies to identify easy training examples. Experimental results show that simple continuation methods are more promising than curriculum learning ones since they allow for slightly improving the performance of state-of-the-art ?-MART models and provide a faster convergence speed.

AB - In this paper we explore the use of Continuation Methods and Curriculum Learning techniques in the area of Learning to Rank. The basic idea is to design the training process as a learning path across increasingly complex training instances and objective functions. We propose to instantiate continuation methods in Learning to Rank by changing the IR measure to optimize during training, and we present two different curriculum learning strategies to identify easy training examples. Experimental results show that simple continuation methods are more promising than curriculum learning ones since they allow for slightly improving the performance of state-of-the-art ?-MART models and provide a faster convergence speed.

KW - Curriculum learning

KW - Lambdamart

KW - Learning to rank

UR - http://www.scopus.com/inward/record.url?scp=85058035587&partnerID=8YFLogxK

U2 - 10.1145/3269206.3269239

DO - 10.1145/3269206.3269239

M3 - Article in proceedings

AN - SCOPUS:85058035587

SP - 1523

EP - 1526

BT - CIKM 2018 - Proceedings of the 27th ACM International Conference on Information and Knowledge Management

A2 - Paton, Norman

A2 - Candan, Selcuk

A2 - Wang, Haixun

A2 - Allan, James

A2 - Agrawal, Rakesh

A2 - Labrinidis, Alexandros

A2 - Cuzzocrea, Alfredo

A2 - Zaki, Mohammed

A2 - Srivastava, Divesh

A2 - Broder, Andrei

A2 - Schuster, Assaf

PB - Association for Computing Machinery, Inc.

T2 - 27th ACM International Conference on Information and Knowledge Management, CIKM 2018

Y2 - 22 October 2018 through 26 October 2018

ER -

Continuation methods and curriculum learning for learning to rank

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this