返回
Operator learning with Gaussian processes
DOI:10.1016/j.cma.2024.117581.png)
摘要
En 中文
Operator learning focuses on approximating mappings c dagger V- V between infinite-dimensional spaces of functions, such as - Rand - R. This makes it particularly suitable for solving parametric nonlinear partial differential equations (PDEs). Recent advancements in machine learning (ML) have brought operator learning to the forefront of research. While most progress in this area has been driven by variants of deep neural networks (NNs), recent studies have demonstrated that Gaussian process (GP)/kernel-based methods can also be competitive. These methods offer advantages in terms of interpretability and provide theoretical and computational guarantees. In this article, we introduce a hybrid GP/NN-based framework for operator learning, leveraging the strengths of both deep neural networks and kernel methods. Instead of directly approximating the function-valued operator c dagger, we use a GP to approximate its associated real-valued bilinear form c dagger V x V*- R. This bilinear form is defined by the dual pairing c dagger ( , ) = [ , c dagger ( )] , which allows us to recover the operator c dagger through c dagger ( )( ) = c dagger ( , ). The mean function of the GP can be set to zero or parameterized by a neural operator and for each setting we develop a robust and scalable training mechanism based on maximum likelihood estimation (MLE) that can optionally leverage the physics involved. Numerical benchmarks demonstrate our method's scope, scalability, efficiency, and robustness; showing that (1) it enhances the performance of a base neural operator by using it as the mean function of a GP, and (2) it enables the construction of zero-shot data-driven models that can make accurate predictions without any prior training. Additionally, our framework (a) naturally extends to cases where c dagger V- & prod; =1 V maps into a vector of functions, and (b) benefits from computational speed-ups achieved through product kernel structures and Kronecker product matrix representations of the underlying kernel matrices.1
Keyword:
Operator learning
Gaussian processes
Neural operators
Zero-shot learning
Optimal recovery
AI总结
对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
7.3
论文数:
1.3W
被引数:
7.5W
机构
引用论文
Multiscale modeling of materials: Computing, data science, uncertainty and goal-oriented optimization材料的多尺度建模: 计算,数据科学,不确定性和面向目标的优化
Wavelet Neural Operator for solving parametric partial differential equations in computational mechanics problems求解计算力学问题中参数偏微分方程的小波神经算子

