Variational Weighting for Kernel Density Ratios
Machine Learning
2023-11-07 v1 Machine Learning
Abstract
Kernel density estimation (KDE) is integral to a range of generative and discriminative tasks in machine learning. Drawing upon tools from the multidimensional calculus of variations, we derive an optimal weight function that reduces bias in standard kernel density estimates for density ratios, leading to improved estimates of prediction posteriors and information-theoretic measures. In the process, we shed light on some fundamental aspects of density estimation, particularly from the perspective of algorithms that employ KDEs as their main building blocks.
Cite
@article{arxiv.2311.03001,
title = {Variational Weighting for Kernel Density Ratios},
author = {Sangwoong Yoon and Frank C. Park and Gunsu S Yun and Iljung Kim and Yung-Kyun Noh},
journal= {arXiv preprint arXiv:2311.03001},
year = {2023}
}
Comments
NeurIPS 2023