English

The $f$-Divergence Expectation Iteration Scheme

Statistics Theory 2021-03-16 v2 Statistics Theory

Abstract

This paper introduces the ff-EI(ϕ)(\phi) algorithm, a novel iterative algorithm which operates on measures and performs ff-divergence minimisation in a Bayesian framework. We prove that for a rich family of values of (f,ϕ)(f,\phi) this algorithm leads at each step to a systematic decrease in the ff-divergence and show that we achieve an optimum. In the particular case where we consider a weighted sum of Dirac measures and the α\alpha-divergence, we obtain that the calculations involved in the ff-EI(ϕ)(\phi) algorithm simplify to gradient-based computations. Empirical results support the claim that the ff-EI(ϕ)(\phi) algorithm serves as a powerful tool to assist Variational methods.

Keywords

Cite

@article{arxiv.1909.12239,
  title  = {The $f$-Divergence Expectation Iteration Scheme},
  author = {Kamélia Daudel and Randal Douc and François Portier and François Roueff},
  journal= {arXiv preprint arXiv:1909.12239},
  year   = {2021}
}

Comments

This content ended up being split into the papers arXiv:2005.10618 and arXiv:2103.05684, which correspond to two separate and more in-depth approaches

R2 v1 2026-06-23T11:27:12.621Z