Projected estimators for robust semi-supervised classification

Jesse H. Krijthe*, Marco Loog

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

15 Citations (Scopus)
48 Downloads (Pure)

Abstract

For semi-supervised techniques to be applied safely in practice we at least want methods to outperform their supervised counterparts. We study this question for classification using the well-known quadratic surrogate loss function. Unlike other approaches to semi-supervised learning, the procedure proposed in this work does not rely on assumptions that are not intrinsic to the classifier at hand. Using a projection of the supervised estimate onto a set of constraints imposed by the unlabeled data, we find we can safely improve over the supervised solution in terms of this quadratic loss. More specifically, we prove that, measured on the labeled and unlabeled training data, this semi-supervised procedure never gives a lower quadratic loss than the supervised alternative. To our knowledge this is the first approach that offers such strong, albeit conservative, guarantees for improvement over the supervised solution. The characteristics of our approach are explicated using benchmark datasets to further understand the similarities and differences between the quadratic loss criterion used in the theoretical results and the classification accuracy typically considered in practice.

Original languageEnglish
Pages (from-to)993-1008
Number of pages16
JournalMachine Learning
Volume106
Issue number7
DOIs
Publication statusPublished - 1 Jul 2017

Keywords

  • Least squares classification
  • Projection
  • Semi-supervised learning

Fingerprint

Dive into the research topics of 'Projected estimators for robust semi-supervised classification'. Together they form a unique fingerprint.

Cite this