arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.01988stat.MLcs.LG
收藏

在希尔伯特空间中的随机插值

Stochastic Interpolants in Hilbert Spaces

Department of Engineering, University of Cambridge, Cambridge, United Kingdom(工程系,剑桥大学,剑桥,英国)

James Boran Yu, RuiKang OuYang, Julien Horwood, José Miguel Hernández-Lobato

AI总结:

本文提出在无限维希尔伯特空间中构建随机插值框架,通过理论支持和实验验证,实现了对复杂函数分布的高效生成,推动了科学发现的应用。

摘要:

尽管扩散模型已成功扩展到函数值数据,但提供灵活方式连接任意分布的随机插值仍局限于有限维设置。本文通过建立在无限维希尔伯特空间中的随机插值的严谨框架来弥合这一差距。我们提供了全面的理论基础,包括对well-posedness的证明和显式误差界。我们展示了所提出框架在条件生成中的有效性,特别关注复杂的基于偏微分方程的基准测试。通过使生成任意功能分布之间的桥梁成为可能,我们的方法实现了最先进的结果,提供了一种强大的、通用的科学发现工具。

英文摘要:

Although diffusion models have successfully extended to function-valued data, stochastic interpolants -- which offer a flexible way to bridge arbitrary distributions -- remain limited to finite-dimensional settings. This work bridges this gap by establishing a rigorous framework for stochastic interpolants in infinite-dimensional Hilbert spaces. We provide comprehensive theoretical foundations, including proofs of well-posedness and explicit error bounds. We demonstrate the effectiveness of the proposed framework for conditional generation, focusing particularly on complex PDE-based benchmarks. By enabling generative bridges between arbitrary functional distributions, our approach achieves state-of-the-art results, offering a powerful, general-purpose tool for scientific discovery.

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。

↑