A Mathematical Guide to Operator Learning

Abstract

Operator learning aims to discover properties of an underlying dynamical system or partial differential equation (PDE) from data. Here, we present a step-by-step guide to operator learning. We explain the types of problems and PDEs amenable to operator learning, discuss various neural network architectures, and explain how to employ numerical PDE solvers effectively. We also give advice on how to create and manage training data and conduct optimization. We offer intuition behind the various neural network architectures employed in operator learning by motivating them from the point-of-view of numerical linear algebra.

45 pages, 11 figures

References in corpus (17)

• A physics-informed variational DeepONet for predicting the crack path in brittle materials
• Reliable extrapolation of deep neural operators informed by physics or sparse observations
• Long-term predictions of turbulence by implicit U-Net enhanced Fourier neural operator
• BINN: A deep learning approach for computational mechanics problems based on boundary integral equations
• Attention-Enhanced Neural Network Models for Turbulence Simulation
• Multi-fidelity prediction of fluid flow and temperature field based on transfer learning using Fourier Neural Operator
• Data-driven discovery of Green's functions with human-understandable deep learning
• Learning Operators with Coupled Attention
• Green's matrices of second order elliptic systems with measurable coefficients in two dimensional domains
• SVD Perspectives for Augmenting DeepONet Flexibility and Interpretability
• Learning elliptic partial differential equations with randomized linear algebra
• PPDONet: Deep Operator Networks for Fast Prediction of Steady-State Solutions in Disk-Planet Systems
• Generic bounds on the approximation error for physics-informed (and) operator learning
• On the Green's matrices of strongly parabolic systems of second order
• Convergence Rates for Learning Linear Operators from Noisy Data
• Elliptic PDE learning is provably data-efficient
• Learning Green's functions associated with time-dependent partial differential equations

Cited by in corpus (4)

• A finite element-based physics-informed operator learning framework for spatiotemporal partial differential equations on arbitrary domains
• A Spectral-based Physics-informed Finite Operator Learning for Prediction of Mechanical Behavior of Microstructures
• An Advanced Physics-Informed Neural Operator for Comprehensive Design Optimization of Highly-Nonlinear Systems: An Aerospace Composites Processing Case Study
• Randomized Strong Recursive Skeletonization: Simultaneous Compression and LU Factorization of Hierarchical Matrices using Matrix-Vector Products