Remarks on kernel Bayes' rule
Machine Learning
2018-04-23 v2
Abstract
Kernel Bayes' rule has been proposed as a nonparametric kernel-based method to realize Bayesian inference in reproducing kernel Hilbert spaces. However, we demonstrate both theoretically and experimentally that the prediction result by kernel Bayes' rule is in some cases unnatural. We consider that this phenomenon is in part due to the fact that the assumptions in kernel Bayes' rule do not hold in general.
Keywords
Cite
@article{arxiv.1507.01059,
title = {Remarks on kernel Bayes' rule},
author = {Hisashi Johno and Kazunori Nakamoto and Tatsuhiko Saigo},
journal= {arXiv preprint arXiv:1507.01059},
year = {2018}
}
Comments
22 pages, 5 figures