AISTATS 2021poster17 citations

A Theoretical Characterization of Semi-supervised Learning with Self-training for Gaussian Mixture Models

Samet Oymak, Talha Cihad Gulcu

Abstract

Self-training is a classical approach in semi-supervised learning which is successfully applied to a variety of machine learning problems. Self-training algorithms generate pseudo-labels for the unlabeled examples and progressively refine these pseudo-labels which hopefully coincides with the actual labels. This work provides theoretical insights into self-training algorithms with a focus on linear classifiers. First, we provide a sample complexity analysis for Gaussian mixture models with two components. This is established by sharp non-asymptotic characterization of the self-training iterations which captures the evolution of the model accuracy in terms of a fixed-point iteration. Our analysis reveals the provable benefits of rejecting samples with low confidence and demonstrates how self-training iterations can gracefully improve the model accuracy. Secondly, we study a generalized GMM where the component means follow a distribution. We demonstrate that ridge regularization and class margin (i.e. separation between the component means) is crucial for the success and lack of regularization may prevent self-training from identifying the core features in the data.

BibTeX
@InProceedings{pmlr-v130-oymak21a,
  title = 	 { A Theoretical Characterization of Semi-supervised Learning with Self-training for Gaussian Mixture Models },
  author =       {Oymak, Samet and Cihad Gulcu, Talha},
  booktitle = 	 {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {3601--3609},
  year = 	 {2021},
  editor = 	 {Banerjee, Arindam and Fukumizu, Kenji},
  volume = 	 {130},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {13--15 Apr},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v130/oymak21a/oymak21a.pdf},
  url = 	 {https://proceedings.mlr.press/v130/oymak21a.html},
  abstract = 	 { Self-training is a classical approach in semi-supervised learning which is successfully applied to a variety of machine learning problems. Self-training algorithms generate pseudo-labels for the unlabeled examples and progressively refine these pseudo-labels which hopefully coincides with the actual labels. This work provides theoretical insights into self-training algorithms with a focus on linear classifiers. First, we provide a sample complexity analysis for Gaussian mixture models with two components. This is established by sharp non-asymptotic characterization of the self-training iterations which captures the evolution of the model accuracy in terms of a fixed-point iteration. Our analysis reveals the provable benefits of rejecting samples with low confidence and demonstrates how self-training iterations can gracefully improve the model accuracy. Secondly, we study a generalized GMM where the component means follow a distribution. We demonstrate that ridge regularization and class margin (i.e. separation between the component means) is crucial for the success and lack of regularization may prevent self-training from identifying the core features in the data. }
}
A Theoretical Characterization of Semi-supervised Learning with Self-training for Gaussian Mixture Models · AISTATS 2021