Unlabeled Data Help in Graph-Based Semi-Supervised Learning: A Bayesian Nonparametrics Perspective

Abstract

In this paper we analyze the graph-based approach to semi-supervised learning under a manifold assumption. We adopt a Bayesian perspective and demonstrate that, for a suitable choice of prior constructed with sufficiently many unlabeled data, the posterior contracts around the truth at a rate that is minimax optimal up to a logarithmic factor. Our theory covers both regression and classification.

Cite

Text

Sanz-Alonso and Yang. "Unlabeled Data Help in Graph-Based Semi-Supervised Learning: A Bayesian Nonparametrics Perspective." Journal of Machine Learning Research, 2022.

Markdown

[Sanz-Alonso and Yang. "Unlabeled Data Help in Graph-Based Semi-Supervised Learning: A Bayesian Nonparametrics Perspective." Journal of Machine Learning Research, 2022.](https://mlanthology.org/jmlr/2022/sanzalonso2022jmlr-unlabeled/)

BibTeX

@article{sanzalonso2022jmlr-unlabeled,
  title     = {{Unlabeled Data Help in Graph-Based Semi-Supervised Learning: A Bayesian Nonparametrics Perspective}},
  author    = {Sanz-Alonso, Daniel and Yang, Ruiyi},
  journal   = {Journal of Machine Learning Research},
  year      = {2022},
  pages     = {1-28},
  volume    = {23},
  url       = {https://mlanthology.org/jmlr/2022/sanzalonso2022jmlr-unlabeled/}
}