Semigroup Kernels on Finite Sets

Abstract

Complex objects can often be conveniently represented by finite sets of simpler components, such as images by sets of patches or texts by bags of words. We study the class of positive definite (p.d.) kernels for two such objects that can be expressed as a function of the merger of their respective sets of components. We prove a general integral representa- tion of such kernels and present two particular examples. One of them leads to a kernel for sets of points living in a space endowed itself with a positive definite kernel. We provide experimental results on a benchmark experiment of handwritten digits image classification which illustrate the validity of the approach.

Cite

Text

Cuturi and Vert. "Semigroup Kernels on Finite Sets." Neural Information Processing Systems, 2004.

Markdown

[Cuturi and Vert. "Semigroup Kernels on Finite Sets." Neural Information Processing Systems, 2004.](https://mlanthology.org/neurips/2004/cuturi2004neurips-semigroup/)

BibTeX

@inproceedings{cuturi2004neurips-semigroup,
  title     = {{Semigroup Kernels on Finite Sets}},
  author    = {Cuturi, Marco and Vert, Jean-philippe},
  booktitle = {Neural Information Processing Systems},
  year      = {2004},
  pages     = {329-336},
  url       = {https://mlanthology.org/neurips/2004/cuturi2004neurips-semigroup/}
}