Tighter Variational Representations of F-Divergences via Restriction to Probability Measures

Abstract

We show that the variational representations for f-divergences currently used in the literature can be tightened. This has implications to a number of methods recently proposed based on this representation. As an example application we use our tighter representation to derive a general f-divergence estimator based on two i.i.d. samples and derive the dual program for this estimator that performs well empirically. We also point out a connection between our estimator and MMD.

Cite

Text

Ruderman et al. "Tighter Variational Representations of F-Divergences via Restriction to Probability Measures." International Conference on Machine Learning, 2012.

Markdown

[Ruderman et al. "Tighter Variational Representations of F-Divergences via Restriction to Probability Measures." International Conference on Machine Learning, 2012.](https://mlanthology.org/icml/2012/ruderman2012icml-tighter/)

BibTeX

@inproceedings{ruderman2012icml-tighter,
  title     = {{Tighter Variational Representations of F-Divergences via Restriction to Probability Measures}},
  author    = {Ruderman, Avraham and Reid, Mark D. and García-García, Dario and Petterson, James},
  booktitle = {International Conference on Machine Learning},
  year      = {2012},
  url       = {https://mlanthology.org/icml/2012/ruderman2012icml-tighter/}
}