arrow
返回

Operator learning with Gaussian processes

delete2025-02-01
delete0
delete
OA
AI
C
Carlos Mora
A
Amin Yousefpour
S
Shirin Hosseinmardi
H
Houman Owhadi
R
Ramin Bostanabad *
DOI:10.1016/j.cma.2024.117581delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

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

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

Computer Methods in Applied Mechanics and Engineering 封面图
Computer Methods in Applied Mechanics and Engineering
IF:
7.3
论文数:
1.3W
被引数:
7.5W

机构

University of California System 封面图
University of California System
学者数:
37.7W
论文数: 33.8W
被引数: 55.6W
U
University of California Irvine
学者数:
934
论文数: 345
被引数: 1.7W
引用论文

引用论文

CAMEA ESS – The continuous angle multi-energy analysis indirect geometry spectrometer for the European Spallation Source
err2015-01-23
err0
errOAAI
errP.G. Freeman; J.O. Birk; M. Markó; M. Bertelsen; J. Larsen; N.B. Christensen; K. Lefmann; J. Jacobsen; Ch. Niedermayer; F. Juranyi; H.M. Ronnow
err分享
err收藏
Multi-fidelity cost-aware Bayesian optimization多保真度成本感知贝叶斯优化
err2023-03-01
err38
errOAAI
errFoumani, Zahra Zanjani; Shishehbor, Mehdi; Yousefpour, Amin; Bostanabad, Ramin
err分享
err收藏
err分享
err收藏
The Fermi Liquid theory of4He in3He
err1992-12-01
err0
PREAI
errD. O. Edwards; M. S. Pettersen; T. G. Culman
err分享
err收藏
err分享
err收藏
Batteries: 3D Hyperbranched Hollow Carbon Nanorod Architectures for High‐Performance Lithium‐Sulfur Batteries (Adv. Energy Mater. 8/2014)
err2014-06-03
err0
PREAI
errShuangqiang Chen; Xiaodan Huang; Hao Liu; Bing Sun; Waikong Yeoh; Kefei Li; Jinqiang Zhang; Guoxiu Wang
err分享
err收藏
err分享
err收藏
err
IF0
err
err0
PREAI
err
err分享
err收藏
学者 查看更多内容