ML Anthology

Improved Graph Laplacian via Geometric Self-Consistency

Dominique Joncas, Marina Meila, James McQueen
NeurIPS 2017 pp. 4457-4466
/neurips/2017/joncas2017neurips-improved/

Abstract

We address the problem of setting the kernel bandwidth, epps, used by Manifold Learning algorithms to construct the graph Laplacian. Exploiting the connection between manifold geometry, represented by the Riemannian metric, and the Laplace-Beltrami operator, we set epps by optimizing the Laplacian's ability to preserve the geometry of the data. Experiments show that this principled approach is effective and robust

PDF NeurIPS Semantic Scholar

Cite

Text

Joncas et al. "Improved Graph Laplacian via Geometric Self-Consistency." Neural Information Processing Systems, 2017.

Markdown

[Joncas et al. "Improved Graph Laplacian via Geometric Self-Consistency." Neural Information Processing Systems, 2017.](https://mlanthology.org/neurips/2017/joncas2017neurips-improved/)

BibTeX

@inproceedings{joncas2017neurips-improved,
  title     = {{Improved Graph Laplacian via Geometric Self-Consistency}},
  author    = {Joncas, Dominique and Meila, Marina and McQueen, James},
  booktitle = {Neural Information Processing Systems},
  year      = {2017},
  pages     = {4457-4466},
  url       = {https://mlanthology.org/neurips/2017/joncas2017neurips-improved/}
}