Iterative Non-Linear Dimensionality Reduction with Manifold Sculpting

Abstract

Many algorithms have been recently developed for reducing dimensionality by projecting data onto an intrinsic non-linear manifold. Unfortunately, existing algo- rithms often lose significant precision in this transformation. Manifold Sculpting is a new algorithm that iteratively reduces dimensionality by simulating surface tension in local neighborhoods. We present several experiments that show Man- ifold Sculpting yields more accurate results than existing algorithms with both generated and natural data-sets. Manifold Sculpting is also able to benefit from both prior dimensionality reduction efforts.

Cite

Text

Gashler et al. "Iterative Non-Linear Dimensionality Reduction with Manifold Sculpting." Neural Information Processing Systems, 2007.

Markdown

[Gashler et al. "Iterative Non-Linear Dimensionality Reduction with Manifold Sculpting." Neural Information Processing Systems, 2007.](https://mlanthology.org/neurips/2007/gashler2007neurips-iterative/)

BibTeX

@inproceedings{gashler2007neurips-iterative,
  title     = {{Iterative Non-Linear Dimensionality Reduction with Manifold Sculpting}},
  author    = {Gashler, Michael and Ventura, Dan and Martinez, Tony},
  booktitle = {Neural Information Processing Systems},
  year      = {2007},
  pages     = {513-520},
  url       = {https://mlanthology.org/neurips/2007/gashler2007neurips-iterative/}
}