On the prediction loss of the lasso in the partially labeled setting

Pierre C. Bellec, Arnak S. Dalalyan, Edwin Grappin, Quentin Paris

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

In this paper we revisit the risk bounds of the lasso estimator in the context of transductive and semi-supervised learning. In other terms, the setting under consideration is that of regression with random design under partial labeling. The main goal is to obtain user-friendly bounds on the off-sample prediction risk. To this end, the simple setting of bounded response variable and bounded (high-dimensional) covariates is considered. We propose some new adaptations of the lasso to these settings and establish oracle inequalities both in expectation and in deviation. These results provide non-asymptotic upper bounds on the risk that highlight the interplay between the bias due to the mis-specification of the linear model, the bias due to the approximate sparsity and the variance. They also demonstrate that the presence of a large number of unlabeled features may have significant positive impact in the situations where the restricted eigenvalue of the design matrix vanishes or is very small.

Original languageEnglish (US)
Pages (from-to)3443-3472
Number of pages30
JournalElectronic Journal of Statistics
Volume12
Issue number2
DOIs
StatePublished - 2018

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Keywords

  • High-dimensional regression
  • Lasso
  • Oracle inequality
  • Semi-supervised learning
  • Sparsity
  • Transductive learning

Fingerprint

Dive into the research topics of 'On the prediction loss of the lasso in the partially labeled setting'. Together they form a unique fingerprint.

Cite this