# A Deep Generative Approach to Stratified Learning

Source: [A Deep Generative Approach to Stratified Learning](<https://www.sota2.com/research/paper/69ddca197d0172140ec79a47>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-15T05:00:11.244Z

Authors: Randy Martinez, Rong Tang, Lizhen Lin

Published: 2026-04-12T13:56:59.000Z

## Abstract

While the manifold hypothesis is widely adopted in modern machine learning, complex data is often better modeled as stratified spaces -- unions of manifolds (strata) of varying dimensions. Stratified learning is challenging due to varying dimensionality, intersection singularities, and lack of efficient models in learning the underlying distributions. We provide a deep generative approach to stratified learning by developing two generative frameworks for learning distributions on stratified spaces. The first is a sieve maximum likelihood approach realized via a dimension-aware mixture of variational autoencoders. The second is a diffusion-based framework that explores the score field structure of a mixture. We establish the convergence rates for learning both the ambient and intrinsic distributions, which are shown to be dependent on the intrinsic dimensions and smoothness of the underlying strata. Utilizing the geometry of the score field, we also establish consistency for estimating the intrinsic dimension of each stratum and propose an algorithm that consistently estimates both the number of strata and their dimensions. Theoretical results for both frameworks provide fundamental insights into the interplay of the underlying geometry, the ambient noise level, and deep generative models. Extensive simulations and real dataset applications, such as molecular dynamics, demonstrate the effectiveness of our methods.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Intrinsic Dimension Estimation on Union of circle and sphere n=10000, sigma\_\*=0](<https://www.sota2.com/research/sota/intrinsic-dimension-estimation-on-union-of-circle-and-sphere-n-10000-sigma_-0.md>): Intrinsic Dimension Estimation; Union of circle and sphere n=10000, sigma\_\*=0; Dim 1 Accuracy: 65.9.

- [Intrinsic dimension estimation on Butane dynamics 500 samples](<https://www.sota2.com/research/sota/intrinsic-dimension-estimation-on-butane-dynamics-500-samples.md>): Intrinsic Dimension Estimation; Butane dynamics 500 samples; Dimension Estimate (Dim 1): 322.

- [Intrinsic Dimension Estimation on 4-manifold mixture noise sigma\_star=0.05](<https://www.sota2.com/research/sota/intrinsic-dimension-estimation-on-4-manifold-mixture-noise-sigma_star-0-05.md>): Intrinsic Dimension Estimation; 4-manifold mixture noise sigma\_star=0.05; Overall Accuracy: 85.74.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.10650>)

[PDF](<https://arxiv.org/pdf/2604.10650.pdf>)
