arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.08077stat.MLcs.LGmath.DSstat.CO
收藏

用于生成式建模的核微分同胚测度匹配

Diffeomorphic Measure Matching with Kernels for Generative Modeling

University of Washington(华盛顿大学);California Institute of Technology(加州理工学院)

Biraj Pandey, Bamdad Hosseini, Pau Batlle, Houman Owhadi

AI总结:

本文提出了一种基于ODEs和RKHSs的概率测度传输通用框架,用于最小散度生成式建模与采样,并提供了理论误差界及广泛的数值实验验证。

摘要:

本文提出了一个概率测度传输的通用框架,旨在实现最小散度生成式建模与采样,该框架利用常微分方程(ODEs)和再生核希尔伯特空间(RKHSs),其灵感来源于微分同胚匹配和图像配准的思想。文中对所提方法进行了理论分析,给出了关于模型复杂度、训练集中样本数量以及模型误设的先验误差界。大量的数值实验进一步突出了该方法的性质、优势与不足,并将其适用性扩展到了条件模拟和推断等其他任务。

英文摘要:

This article presents a general framework for the transport of probability measures towards minimum divergence generative modeling and sampling using ordinary differential equations (ODEs) and Reproducing Kernel Hilbert Spaces (RKHSs), inspired by ideas from diffeomorphic matching and image registration. A theoretical analysis of the proposed method is presented, giving a priori error bounds in terms of the complexity of the model, the number of samples in the training set, and model misspecification. An extensive suite of numerical experiments further highlights the properties, strengths, and weaknesses of the method and extends its applicability to other tasks, such as conditional simulation and inference.

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。

↑