Two generative frameworks (sieve MLE via dimension-aware VAE mixtures and diffusion on score fields) are developed for distributions on stratified spaces, with convergence rates depending on intrinsic dimensions and smoothness, plus consistent estimators for the number and dimensions of strata.
SinceGkj(z)andv kj(z) are polynomials with degree at most⌊βk⌋and⌊α k⌋respectively, we can rewrite(A ∗ 2)and (B∗ 2)as (A∗
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Deep Generative Approach to Stratified Learning
Two generative frameworks (sieve MLE via dimension-aware VAE mixtures and diffusion on score fields) are developed for distributions on stratified spaces, with convergence rates depending on intrinsic dimensions and smoothness, plus consistent estimators for the number and dimensions of strata.