Limitations of “Limitations of Bayesian Leave-one-out Cross-Validation for Model Selection”

被引：2

作者：

Vehtari A. ^{[1
]}

Simpson D.P. ^{[2
]}

Yao Y. ^{[3
]}

Gelman A. ^{[4
]}

机构：

[1] Department of Computer Science, Aalto University, Espoo

[2] Department of Statisitcs, University of Toronto, Toronto

[3] Department of Statistics, Columbia University, New York, NY

[4] Department of Statistics and Department of Political Science, Columbia University, New York, NY

来源：

Computational Brain & Behavior | 2019年 / 2卷 / 1期

基金：

加拿大自然科学与工程研究理事会; 芬兰科学院; 美国国家科学基金会;

关键词：

M-closed; M-open; Principle of complexity; Reality; Statistical convenience;

D O I：

10.1007/s42113-018-0020-6

中图分类号：

学科分类号：

摘要：

In an earlier article in this journal, Gronau and Wagenmakers (2018) discuss some problems with leave-one-out cross-validation (LOO) for Bayesian model selection. However, the variant of LOO that Gronau and Wagenmakers discuss is at odds with a long literature on how to use LOO well. In this discussion, we discuss the use of LOO in practical data analysis, from the perspective that we need to abandon the idea that there is a device that will produce a single-number decision rule. © 2019, The Author(s).

引用

页码：22 / 27

页数：5

共 50 条

[1] Limitations of Bayesian Leave-One-Out Cross-Validation for Model Selection
Gronau Q.F.
Wagenmakers E.-J.
[J]. Computational Brain & Behavior, 2019, 2 (1) : 1 - 11
[2] Rejoinder: More Limitations of Bayesian Leave-One-Out Cross-Validation
Gronau Q.F.
Wagenmakers E.-J.
[J]. Computational Brain & Behavior, 2019, 2 (1) : 35 - 47
[3] Bayesian Leave-One-Out Cross-Validation for Large Data
Magnusson, Mans
Andersen, Michael Riis
Jonasson, Johan
Vehtari, Aki
[J]. INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 97, 2019, 97
[4] Leave-One-Out Cross-Validation for Bayesian Model Comparison in Large Data
Magnusson, Mans
Andersen, Michael Riis
Jonasson, Johan
Vehtari, Aki
[J]. INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 108, 2020, 108 : 341 - 350
[5] Leave-One-Out Cross-Validation Based Model Selection for Manifold Regularization
Yuan, Jin
Li, Yan-Ming
Liu, Cheng-Liang
Zha, Xuan F.
[J]. ADVANCES IN NEURAL NETWORKS - ISNN 2010, PT 1, PROCEEDINGS, 2010, 6063 : 457 - +
[6] On the Application of Bayesian Leave-one-out Cross-validation to Exoplanet Atmospheric Analysis
Welbanks, Luis
McGill, Peter
Line, Michael
Madhusudhan, Nikku
[J]. ASTRONOMICAL JOURNAL, 2023, 165 (03):
[7] Erratum to: Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC
Aki Vehtari
Andrew Gelman
Jonah Gabry
[J]. Statistics and Computing, 2017, 27 : 1433 - 1433
[8] Robust Leave-One-Out Cross-Validation for High-Dimensional Bayesian Models
Silva, Luca Alessandro
Zanella, Giacomo
[J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2023,
[9] Leave-one-out cross-validation is risk consistent for lasso
Darren Homrighausen
Daniel J. McDonald
[J]. Machine Learning, 2014, 97 : 65 - 78
[10] Leave-one-out cross-validation is risk consistent for lasso
Homrighausen, Darren
McDonald, Daniel J.
[J]. MACHINE LEARNING, 2014, 97 (1-2) : 65 - 78

← 1 2 3 4 5 →