For two-class Gaussian mixtures in high dimensions, optimally regularized maximum likelihood nearly matches Bayes-optimal performance in estimation and prediction error when unlabeled data are abundant.
Asymptotic Bayes risk for Gaussian mixture in a semi- supervised setting
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Analysis of High-dimensional Gaussian Labeled-unlabeled Mixture Model via Message-passing Algorithm
For two-class Gaussian mixtures in high dimensions, optimally regularized maximum likelihood nearly matches Bayes-optimal performance in estimation and prediction error when unlabeled data are abundant.