arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.20979stat.MLcs.LGmath.STstat.TH
收藏

面向组特异性扩散模型的两阶段分数估计

Two-Stage Score Estimation for Group-Specific Diffusion Models

University of California, Los Angeles(加州大学洛杉矶分校)

Huitao Yang, Oscar Hernan Madrid Padilla

AI总结:

该研究针对目标分布观测有限的生成建模问题,提出两阶段扩散迁移统计框架,推导收敛速率并通过模拟与图像基准验证其有效性。

摘要:

跨相关分布的生成建模提出了一个问题:共享结构如何在目标分布观测有限的情况下支持准确生成?我们开发了一种用于扩散迁移的统计框架,该框架分析共享的池化分数以及目标特定的偏移量。在池化密度的Besov正则性和偏移量的稳定分层组合模型下,我们推导了收敛速率,这些速率分离了学习这些组件的成本,并确定了何时更简单的偏移量比单独拟合能产生更快的风险边界速率。在适当的边界和正则性假设下,我们证明当扩散时间趋近于零时,偏移量保持有界且联合正则。这些结果允许较低的时间截断以多项式方式消失,并为扩散模型作为密度估计器提供总变差保证。合成模拟和图像基准为该框架提供了概念验证。

英文摘要:

Generative modeling across related distributions raises a question: how can shared structure support accurate generation for a target distribution with limited observations? We develop a statistical framework for diffusion transfer that analyzes a shared pooled score together with a target-specific offset. Under Besov regularity of the pooled density and a stable hierarchical composition model for the offset, we derive convergence rates that separate the costs of learning these components and identify when a simpler offset yields a faster risk-bound rate than separate fitting. Under suitable boundary and regularity assumptions, we prove that the offset remains bounded and jointly regular as diffusion time approaches zero. These results allow the lower time cutoff to vanish polynomially and provide total-variation guarantees for diffusion models as density estimators. Synthetic simulations and image benchmarks provide a proof of concept for the framework.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑