The Pessimistic Limits and Possibilities of Margin-Based Losses in Semi-Supervised Learning

Abstract

Consider a classification problem where we have both labeled and unlabeled data available. We show that for linear classifiers defined by convex margin-based surrogate losses that are decreasing, it is impossible to construct \emph{any} semi-supervised approach that is able to guarantee an improvement over the supervised classifier measured by this surrogate loss on the labeled and unlabeled data. For convex margin-based loss functions that also increase, we demonstrate safe improvements \emph{are} possible.

Cite

Text

Krijthe and Loog. "The Pessimistic Limits and Possibilities of Margin-Based Losses in Semi-Supervised Learning." Neural Information Processing Systems, 2018.

Markdown

[Krijthe and Loog. "The Pessimistic Limits and Possibilities of Margin-Based Losses in Semi-Supervised Learning." Neural Information Processing Systems, 2018.](https://mlanthology.org/neurips/2018/krijthe2018neurips-pessimistic/)

BibTeX

@inproceedings{krijthe2018neurips-pessimistic,
  title     = {{The Pessimistic Limits and Possibilities of Margin-Based Losses in Semi-Supervised Learning}},
  author    = {Krijthe, Jesse and Loog, Marco},
  booktitle = {Neural Information Processing Systems},
  year      = {2018},
  pages     = {1790-1799},
  url       = {https://mlanthology.org/neurips/2018/krijthe2018neurips-pessimistic/}
}