A robust Bayesian estimation approach for the imprecise Plackett-Luce model

Tathagata Basu; Sébastien Destercke; Benjamin Quost

doi:10.1007/978-3-031-08971-8_61

A robust Bayesian estimation approach for the imprecise Plackett-Luce model

Tathagata Basu^*, Sébastien Destercke, Benjamin Quost

^*Corresponding author for this work

Civil And Environmental Engineering

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

Abstract

Learning to rank has become an important part in the fields of machine learning and statistical learning. Rankings are indeed present in many applications, including cognitive psychology, recommender systems, sports tournament or automated algorithm choices. Rankings are however prone to subjectivity (when provided by users) and to incompleteness (when a contestant is missing, or users only report partial preferences). Robust or cautious approaches may overcome such issues. In this paper, we develop a Bayesian robust approach for a commonly used parametric model, the Plackett-Luce (PL) model. This allows us to obtain interval-valued parameter estimates for the strength parameter of the Plackett-Luce model. We illustrate our method with both synthetic and real data to show the usefulness of skeptic inference.

Original language	English
Title of host publication	Information Processing and Management of Uncertainty in Knowledge-Based Systems
Subtitle of host publication	19th International Conference, IPMU 2022, Proceedings
Editors	Davide Ciucci, Inés Couso, Jesús Medina, Dominik Ślęzak, Davide Petturiti, Bernadette Bouchon-Meunier, Ronald R. Yager
Place of Publication	Cham, Switzerland
Publisher	Springer
Pages	757-769
Number of pages	13
ISBN (Electronic)	9783031089718
ISBN (Print)	9783031089701
DOIs	https://doi.org/10.1007/978-3-031-08971-8_61
Publication status	Published - 4 Jul 2022
Event	19th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2022 - Milan, Italy Duration: 11 Jul 2022 → 15 Jul 2022

Publication series

Name	Communications in Computer and Information Science
Volume	1601 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	19th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2022
Country/Territory	Italy
City	Milan
Period	11/07/22 → 15/07/22

Funding

This research was funded by the project PreServe (ANR Grant ANR-18-CE23-0008).

Keywords

Bayesian analysis
imprecise probability
Plackett-Luce model
preference learning

Access to Document

10.1007/978-3-031-08971-8_61

Cite this

Basu, T., Destercke, S., & Quost, B. (2022). A robust Bayesian estimation approach for the imprecise Plackett-Luce model. In D. Ciucci, I. Couso, J. Medina, D. Ślęzak, D. Petturiti, B. Bouchon-Meunier, & R. R. Yager (Eds.), Information Processing and Management of Uncertainty in Knowledge-Based Systems: 19th International Conference, IPMU 2022, Proceedings (pp. 757-769). (Communications in Computer and Information Science; Vol. 1601 CCIS). Springer. https://doi.org/10.1007/978-3-031-08971-8_61

Basu, Tathagata ; Destercke, Sébastien ; Quost, Benjamin. / A robust Bayesian estimation approach for the imprecise Plackett-Luce model. Information Processing and Management of Uncertainty in Knowledge-Based Systems: 19th International Conference, IPMU 2022, Proceedings. editor / Davide Ciucci ; Inés Couso ; Jesús Medina ; Dominik Ślęzak ; Davide Petturiti ; Bernadette Bouchon-Meunier ; Ronald R. Yager. Cham, Switzerland : Springer, 2022. pp. 757-769 (Communications in Computer and Information Science).

@inproceedings{baf9804f8abf4c9d8164a1ccf81269e3,

title = "A robust Bayesian estimation approach for the imprecise Plackett-Luce model",

abstract = "Learning to rank has become an important part in the fields of machine learning and statistical learning. Rankings are indeed present in many applications, including cognitive psychology, recommender systems, sports tournament or automated algorithm choices. Rankings are however prone to subjectivity (when provided by users) and to incompleteness (when a contestant is missing, or users only report partial preferences). Robust or cautious approaches may overcome such issues. In this paper, we develop a Bayesian robust approach for a commonly used parametric model, the Plackett-Luce (PL) model. This allows us to obtain interval-valued parameter estimates for the strength parameter of the Plackett-Luce model. We illustrate our method with both synthetic and real data to show the usefulness of skeptic inference.",

keywords = "Bayesian analysis, imprecise probability, Plackett-Luce model, preference learning",

author = "Tathagata Basu and S{\'e}bastien Destercke and Benjamin Quost",

year = "2022",

month = jul,

day = "4",

doi = "10.1007/978-3-031-08971-8\_61",

language = "English",

isbn = "9783031089701",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "757--769",

editor = "Davide Ciucci and In{\'e}s Couso and Jes{\'u}s Medina and Dominik {\'S}l{\c e}zak and Davide Petturiti and Bernadette Bouchon-Meunier and Yager, \{Ronald R.\}",

booktitle = "Information Processing and Management of Uncertainty in Knowledge-Based Systems",

address = "Switzerland",

note = "19th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2022 ; Conference date: 11-07-2022 Through 15-07-2022",

}

Basu, T, Destercke, S & Quost, B 2022, A robust Bayesian estimation approach for the imprecise Plackett-Luce model. in D Ciucci, I Couso, J Medina, D Ślęzak, D Petturiti, B Bouchon-Meunier & RR Yager (eds), Information Processing and Management of Uncertainty in Knowledge-Based Systems: 19th International Conference, IPMU 2022, Proceedings. Communications in Computer and Information Science, vol. 1601 CCIS, Springer, Cham, Switzerland, pp. 757-769, 19th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2022, Milan, Italy, 11/07/22. https://doi.org/10.1007/978-3-031-08971-8_61

A robust Bayesian estimation approach for the imprecise Plackett-Luce model. / Basu, Tathagata; Destercke, Sébastien; Quost, Benjamin.
Information Processing and Management of Uncertainty in Knowledge-Based Systems: 19th International Conference, IPMU 2022, Proceedings. ed. / Davide Ciucci; Inés Couso; Jesús Medina; Dominik Ślęzak; Davide Petturiti; Bernadette Bouchon-Meunier; Ronald R. Yager. Cham, Switzerland: Springer, 2022. p. 757-769 (Communications in Computer and Information Science; Vol. 1601 CCIS).

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

TY - GEN

T1 - A robust Bayesian estimation approach for the imprecise Plackett-Luce model

AU - Basu, Tathagata

AU - Destercke, Sébastien

AU - Quost, Benjamin

PY - 2022/7/4

Y1 - 2022/7/4

N2 - Learning to rank has become an important part in the fields of machine learning and statistical learning. Rankings are indeed present in many applications, including cognitive psychology, recommender systems, sports tournament or automated algorithm choices. Rankings are however prone to subjectivity (when provided by users) and to incompleteness (when a contestant is missing, or users only report partial preferences). Robust or cautious approaches may overcome such issues. In this paper, we develop a Bayesian robust approach for a commonly used parametric model, the Plackett-Luce (PL) model. This allows us to obtain interval-valued parameter estimates for the strength parameter of the Plackett-Luce model. We illustrate our method with both synthetic and real data to show the usefulness of skeptic inference.

AB - Learning to rank has become an important part in the fields of machine learning and statistical learning. Rankings are indeed present in many applications, including cognitive psychology, recommender systems, sports tournament or automated algorithm choices. Rankings are however prone to subjectivity (when provided by users) and to incompleteness (when a contestant is missing, or users only report partial preferences). Robust or cautious approaches may overcome such issues. In this paper, we develop a Bayesian robust approach for a commonly used parametric model, the Plackett-Luce (PL) model. This allows us to obtain interval-valued parameter estimates for the strength parameter of the Plackett-Luce model. We illustrate our method with both synthetic and real data to show the usefulness of skeptic inference.

KW - Bayesian analysis

KW - imprecise probability

KW - Plackett-Luce model

KW - preference learning

U2 - 10.1007/978-3-031-08971-8_61

DO - 10.1007/978-3-031-08971-8_61

M3 - Conference contribution book

AN - SCOPUS:85135072198

SN - 9783031089701

T3 - Communications in Computer and Information Science

SP - 757

EP - 769

BT - Information Processing and Management of Uncertainty in Knowledge-Based Systems

A2 - Ciucci, Davide

A2 - Couso, Inés

A2 - Medina, Jesús

A2 - Ślęzak, Dominik

A2 - Petturiti, Davide

A2 - Bouchon-Meunier, Bernadette

A2 - Yager, Ronald R.

PB - Springer

CY - Cham, Switzerland

T2 - 19th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2022

Y2 - 11 July 2022 through 15 July 2022

ER -

Basu T, Destercke S, Quost B. A robust Bayesian estimation approach for the imprecise Plackett-Luce model. In Ciucci D, Couso I, Medina J, Ślęzak D, Petturiti D, Bouchon-Meunier B, Yager RR, editors, Information Processing and Management of Uncertainty in Knowledge-Based Systems: 19th International Conference, IPMU 2022, Proceedings. Cham, Switzerland: Springer. 2022. p. 757-769. (Communications in Computer and Information Science). doi: 10.1007/978-3-031-08971-8_61

A robust Bayesian estimation approach for the imprecise Plackett-Luce model

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Fingerprint

Cite this