arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.12902cs.LG
收藏

用于函数不确定性量化的概率神经算子

Probabilistic neural operators for functional uncertainty quantification

Ludwig-Maximilians-Universität München(慕尼黑大学);Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心);University of Tromsø(特罗姆瑟大学);DLR-German Aerospace Center(德国航空航天中心)

Christopher Bülte, Philipp Scholl, Gitta Kutyniok

AI总结:

本文提出概率神经算子(PNO)框架,通过严格适当评分规则的生成建模将不确定性直接整合到训练中,在最小调整现有架构下提升复杂动力系统的不确定性量化与预测校准性能。

摘要:

神经算子旨在纯粹从数据中近似微分方程系统的解算子。它们在跨领域建模复杂动力系统方面取得了巨大成功。然而,模型和数据中固有的不确定性发生至今很少被考虑——这在天气预报等复杂混沌系统中是一个关键局限。在本文中,我们引入了概率神经算子(PNO),这是一个用于学习神经算子输出函数空间上概率分布的框架。PNO通过基于严格适当评分规则的生成建模扩展了神经算子,将不确定性信息直接整合到训练过程中。我们为该方法提供了理论依据,并展示了在不同领域和不同基线下量化不确定性的性能提升。此外,PNO对现有架构只需最小调整,在大多数概率预测任务中表现出提升的性能,并且即使对于长动力轨迹也能产生校准良好的预测分布和充分的不确定性表示。将我们的方法实现到物理应用的大规模模型中,可以改进相应的不确定性量化和极端事件识别,最终促进对此类代理模型预测的更深入理解。

英文摘要:

Neural operators aim to approximate the solution operator of a system of differential equations purely from data. They have shown immense success in modeling complex dynamical systems across various domains. However, the occurrence of uncertainties inherent in both model and data has so far rarely been taken into account\textemdash{}a critical limitation in complex, chaotic systems such as weather forecasting. In this paper, we introduce the probabilistic neural operator (PNO), a framework for learning probability distributions over the output function space of neural operators. PNO extends neural operators with generative modeling based on strictly proper scoring rules, integrating uncertainty information directly into the training process. We provide a theoretical justification for the approach and demonstrate improved performance in quantifying uncertainty across different domains and with respect to different baselines. Furthermore, PNO requires minimal adjustment to existing architectures, shows improved performance for most probabilistic prediction tasks, and leads to well-calibrated predictive distributions and adequate uncertainty representations even for long dynamical trajectories. Implementing our approach into large-scale models for physical applications can lead to improvements in corresponding uncertainty quantification and extreme event identification, ultimately leading to a deeper understanding of the prediction of such surrogate models.

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。

↑