English

Minimizing $f$-Divergences by Interpolating Velocity Fields

Machine Learning 2024-06-07 v3 Machine Learning

Abstract

Many machine learning problems can be seen as approximating a \textit{target} distribution using a \textit{particle} distribution by minimizing their statistical discrepancy. Wasserstein Gradient Flow can move particles along a path that minimizes the ff-divergence between the target and particle distributions. To move particles, we need to calculate the corresponding velocity fields derived from a density ratio function between these two distributions. Previous works estimated such density ratio functions and then differentiated the estimated ratios. These approaches may suffer from overfitting, leading to a less accurate estimate of the velocity fields. Inspired by non-parametric curve fitting, we directly estimate these velocity fields using interpolation techniques. We prove that our estimators are consistent under mild conditions. We validate their effectiveness using novel applications on domain adaptation and missing data imputation.

Keywords

Cite

@article{arxiv.2305.15577,
  title  = {Minimizing $f$-Divergences by Interpolating Velocity Fields},
  author = {Song Liu and Jiahao Yu and Jack Simons and Mingxuan Yi and Mark Beaumont},
  journal= {arXiv preprint arXiv:2305.15577},
  year   = {2024}
}

Comments

This manuscript is an extended version of the ICML2024 version. The code for reproducing our results can be found at https://github.com/anewgithubname/gradest2

R2 v1 2026-06-28T10:45:17.733Z