A spectral framework converts KL divergence estimation into least-squares problems and yields closed-form estimators for relative log-densities and divergences from feature moments.
Enhanced feature learning via regularisation: Integrating neural networks and kernel methods.Journal of Machine Learning Research, 26(172):1–56, 2025.(cited on page 7)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Spectral Framework for Closed-Form Relative Density Estimation
A spectral framework converts KL divergence estimation into least-squares problems and yields closed-form estimators for relative log-densities and divergences from feature moments.