A Non-parametric Fisher Kernel

  1. Pau Figuera 1
  2. Pablo García Bringas 1
  1. 1 Universidad de Deusto
    info

    Universidad de Deusto

    Bilbao, España

    ROR https://ror.org/00ne6sr39

Book:
Hybrid Artificial Intelligent Systems: 16th International Conference, HAIS 2021. Bilbao, Spain. September 22–24, 2021. Proceedings
  1. Hugo Sanjurjo González (coord.)
  2. Iker Pastor López (coord.)
  3. Pablo García Bringas (coord.)
  4. Héctor Quintián (coord.)
  5. Emilio Corchado (coord.)

Publisher: Springer International Publishing AG

ISBN: 978-3-030-86271-8,  978-3-030-86270-1

Year of publication: 2021

Pages: 448-459

Congress: Hybrid Artificial Intelligent Systems (HAIS) (16. 2021. Bilbao)

Type: Conference paper

DIALNET GOOGLE SCHOLAR

Abstract

In this manuscript, we derive a non-parametric version of the Fisher kernel. We obtain this original result from the Non-negative Matrix Factorization with the Kullback-Leibler divergence. By imposing suitable normalization conditions on the obtained factorization, it can be assimilated to a mixture of densities, with no assumptions on the distribution of the parameters. The equivalence between the Kullback- Leibler divergence and the log-likelihood leads to kernelization by simply taking partial derivatives. The estimates provided by this kernel, retain the consistency of the Fisher kernel.

Data source: Dialnet