Multi-Manifold Semi-Supervised Learning
Abstract
We study semi-supervised learning when the data consists of multiple intersecting manifolds. We give a finite sample analysis to quantify the potential gain of using unlabeled data in this multi-manifold setting. We then propose a semi-supervised learning algorithm that separates different manifolds into decision sets, and performs supervised learning within each set. Our algorithm involves a novel application of Hellinger distance and size-constrained spectral clustering. Experiments demonstrate the benefit of our multi-manifold semi-supervised learning approach.
Cite
Text
Goldberg et al. "Multi-Manifold Semi-Supervised Learning." Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics, 2009.Markdown
[Goldberg et al. "Multi-Manifold Semi-Supervised Learning." Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics, 2009.](https://mlanthology.org/aistats/2009/goldberg2009aistats-multimanifold/)BibTeX
@inproceedings{goldberg2009aistats-multimanifold,
title = {{Multi-Manifold Semi-Supervised Learning}},
author = {Goldberg, Andrew and Zhu, Xiaojin and Singh, Aarti and Xu, Zhiting and Nowak, Robert},
booktitle = {Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics},
year = {2009},
pages = {169-176},
volume = {5},
url = {https://mlanthology.org/aistats/2009/goldberg2009aistats-multimanifold/}
}