arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.11794cs.LG
收藏

通过维纳混沌学习隐变量的随机偏微分方程

Latent-Variable Learning of SPDEs via Wiener Chaos

Department of mathematics and physics(数学与物理系);Linnaeus University(林奈大学);physics(物理);Växjö, Sweden(瑞典瓦克斯约)

Sebastian Zeng, Andreas Petersson, Wolfgang Bock

AI总结:

本文提出了一种通过维纳混沌学习隐变量的随机偏微分方程方法,结合谱伽辽金投影与截断维纳混沌展开,实现对随机驱动的动力学建模,从而在无需显式噪声观测的情况下恢复随机结构。

摘要:

我们研究了从时空观测中学习线性随机偏微分方程(SPDEs)的分布的问题。现有的深度学习方法要么假设可以获取驱动噪声或初始条件,要么依赖于确定性替代模型,这些模型无法捕捉内在的随机性。我们提出了一种结构化的隐变量公式,仅需要解决方案实现实观测,并学习底层随机驱动的动力学。我们的方法结合了谱伽辽金投影与截断维纳混沌展开,从而在确定性演化和随机驱动之间实现原则性的分离。这将无限维的SPDE减少为一个参数化的有限系统,该系统 governing 隐变量的时间动力学。隐变量动力学和随机驱动通过变分学习联合推断,允许在训练过程中无需显式观察或模拟噪声即可恢复随机结构。在合成数据上的实证评估表明,在可比较的建模假设下,对于有界和无界的单维空间域,性能处于最先进的水平。

英文摘要:

We study the problem of learning the law of linear stochastic partial differential equations (SPDEs) with additive Gaussian forcing from spatiotemporal observations. Most existing deep learning approaches either assume access to the driving noise or initial condition, or rely on deterministic surrogate models that fail to capture intrinsic stochasticity. We propose a structured latent-variable formulation that requires only observations of solution realizations and learns the underlying randomly forced dynamics. Our approach combines a spectral Galerkin projection with a truncated Wiener chaos expansion, yielding a principled separation between deterministic evolution and stochastic forcing. This reduces the infinite-dimensional SPDE to a finite system of parametrized ordinary differential equations governing latent temporal dynamics. The latent dynamics and stochastic forcing are jointly inferred through variational learning, allowing recovery of stochastic structure without explicit observation or simulation of noise during training. Empirical evaluation on synthetic data demonstrates state-of-the-art performance under comparable modeling assumptions across bounded and unbounded one-dimensional spatial domains.

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。

↑