Linear predictor on linearly-generated data with missing values: non consistency and solutions - Archive ouverte HAL
Conference Papers Year : 2020

Linear predictor on linearly-generated data with missing values: non consistency and solutions

Abstract

We consider building predictors when the data have missing values. We study the seemingly-simple case where the target to predict is a linear function of the fully-observed data and we show that, in the presence of missing values, the optimal predictor may not be linear. In the particular Gaussian case, it can be written as a linear function of multiway interactions between the observed data and the various missing-value indicators. Due to its intrinsic complexity, we study a simple approximation and prove generalization bounds with finite samples, highlighting regimes for which each method performs best. We then show that multilayer perceptrons with ReLU activation functions can be consistent, and can explore good trade-offs between the true model and approximations. Our study highlights the interesting family of models that are beneficial to fit with missing values depending on the amount of data available.
Fichier principal
Vignette du fichier
aistats.pdf (6.88 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-02464569 , version 1 (03-02-2020)
hal-02464569 , version 2 (06-05-2020)

Identifiers

Cite

Marine Le Morvan, Nicolas Prost, Julie Josse, Erwan Scornet, Gaël Varoquaux. Linear predictor on linearly-generated data with missing values: non consistency and solutions. AISTATS 2020 - International Conference on Artificial Intelligence and Statistics, Aug 2020, Online, France. pp.3165-3174. ⟨hal-02464569v2⟩
598 View
217 Download

Altmetric

Share

More