EigenBayes combines spectral estimation of latent factors with adaptive empirical Bayes hyperparameter calibration to shrink superfluous components in overfitted factor models, delivering tractable posteriors and favorable asymptotics.
Overfitted high-dimensional matrix fac- torizations via adaptive spectral shrinkage
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Overfitted high-dimensional matrix factorizations via adaptive spectral shrinkage
EigenBayes combines spectral estimation of latent factors with adaptive empirical Bayes hyperparameter calibration to shrink superfluous components in overfitted factor models, delivering tractable posteriors and favorable asymptotics.