Physics-Informed Neural Operator for Learning Partial Differential Equations
2021/11/06 by Zongyi Li, Li, Zongyi, Hongkai Zheng +13 · 138 citations
Engineering · Physics and Astronomy · #FOS: Computer and information sciences #FOS: Mathematics #Fluid Dynamics and Turbulent Flows #Machine Learning (cs.LG) #Model Reduction and Neural Networks #Nuclear Engineering Thermal-Hydraulics #Numerical Analysis (math.NA)
paper · pdf · doi:10.48550/arxiv.2111.03794
openalex publication_date 2021/11/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Abstract
In this paper, we propose physics-informed neural operators (PINO) that combine training data and physics constraints to learn the solution operator of a given family of parametric Partial Differential Equations (PDE). PINO is the first hybrid approach incorporating data and PDE constraints at different resolutions to learn the operator. Specifically, in PINO, we combine coarse-resolution training data with PDE constraints imposed at a higher resolution. The resulting PINO model can accurately approximate the ground-truth solution operator for many popular PDE families and shows no degradation in accuracy even under zero-shot super-resolution, i.e., being able to predict beyond the resolution of training data. PINO uses the Fourier neural operator (FNO) framework that is guaranteed to be a universal approximator for any continuous operator and discretization-convergent in the limit of mesh refinement. By adding PDE constraints to FNO at a higher resolution, we obtain a high-fidelity reconstruction of the ground-truth operator. Moreover, PINO succeeds in settings where no training data is available and only PDE constraints are imposed, while previous approaches, such as the Physics-Informed Neural Network (PINN), fail due to optimization challenges, e.g., in multi-scale dynamic systems such as Kolmogorov flows.
Citations
Cited by
- PI-MFM: Physics-informed multimodal foundation model for solving partial differential equations
- Global polynomial-time estimation in statistical nonlinear inverse problems via generalized stability
- Extending Fourier Neural Operators for Modeling Parameterized and Coupled PDEs
- Forward and Inverse Mantle Convection with Neural Operators
- Operator learning for models of tear film breakup
- Data-Free Asymptotics-Informed Operator Networks for Singularly Perturbed PDEs
- MAD-NG: Meta-Auto-Decoder Neural Galerkin Method for Solving Parametric Partial Differential Equations
- Variationally correct operator learning: Reduced basis neural operator with a posteriori error estimation
- Solving Functional PDEs with Gaussian Processes and Applications to Functional Renormalization Group Equations
- GeoTransolver: Learning Physics on Irregular Domains Using Multi-scale Geometry Aware Physics Attention Transformer
- Learning Generalizable Neural Operators for Inverse Problems
- Physically consistent model learning for reaction-diffusion systems
- Integrating Fourier Neural Operator with Diffusion Model for Autoregressive Predictions of Three-dimensional Turbulence
- Generalized Spherical Neural Operators: Green's Function Formulation
- FNOPT: Resolution-Agnostic, Self-Supervised Cloth Simulation using Meta-Optimization with Fourier Neural Operators
- PRISMA: Improving the Accuracy-Latency Frontier of Diffusion-based PDE Solvers Using Physics-Informed Spectral Attention
- PIANO: Physics-informed Dual Neural Operator for Precipitation Nowcasting
- Breaking Scale Anchoring: Frequency Representation Learning for Accurate High-Resolution Inference from Low-Resolution Training
- Operator Learning at Machine Precision
- Walrus: A Cross-Domain Foundation Model for Continuum Dynamics
- Physics-Informed Neural Networks for Nonlinear Output Regulation
- Method of Manufactured Learning for Solver-free Training of Neural Operators
- Real-time distortion prediction in metallic additive manufacturing via a physics-informed neural operator approach
- Generalizing PDE Emulation with Equation-Aware Neural Operators
- Physics-Informed Neural Operators for Cardiac Electrophysiology
- Neural-Initialized Newton: Accelerating Nonlinear Finite Elements via Operator Learning
- Event-driven physics-informed operator learning for reliability analysis
- Comparing EPGP Surrogates and Finite Elements Under Degree-of-Freedom Parity
- Learning Under Laws: A Constraint-Projected Neural PDE Solver that Eliminates Hallucinations
- A unified physics-informed generative operator framework for general inverse problems
- Fourier Neural Operators for Two-Phase, 2D Mold-Filling Problems Related to Metal Casting
- A Physics-Informed Neural Operator for Thermal Ranking of Low-Cost Wall Materials in Hot-Dry Climates
- Can Data-Driven Dynamics Reveal Hidden Physics? There Is A Need for Interpretable Neural Operators
- Mixture-of-Experts Operator Transformer for Large-Scale PDE Pre-Training
- Enforcing boundary conditions for physics-informed neural operators
- Reanimating the past: From historical collections of the placenta and uterus to modern imaging, machine learning, and multiscale modeling
- Neural Emulator Superiority: When Machine Learning for PDEs Surpasses its Training Data
- Sparse surface pressure-based reconstruction of the flow around a thick airfoil over a range of angles of attack
- Simultaneously Solving Infinitely Many LQ Mean Field Games In Hilbert Spaces: The Power of Neural Operators
- Infinite Neural Operators: Gaussian processes on functions
- Deep Neural Networks Inspired by Differential Equations
- A physics-aware deep learning model for shear band formation around collapsing pores in shocked reactive materials
- Diagnosing Failure Modes of Neural Operators Across Diverse PDE Families
- The False Promise of Zero-Shot Super-Resolution in Machine-Learned Operators
- A comprehensive comparison of neural operators for 3D industry-scale engineering designs
- Physics-informed Neural-operator Predictive Control for Drag Reduction in Turbulent Flows
- Physics-Informed Neural Controlled Differential Equations for Scalable Long Horizon Multi-Agent Motion Forecasting
- Machine Learning Workflows in Climate Modeling: Design Patterns and Insights from Case Studies
- Trustworthy AI in numerics: On verification algorithms for neural network-based PDE solvers
- Towards Generalizable PDE Dynamics Forecasting via Physics-Guided Invariant Learning
- MORPH: PDE Foundation Models with Arbitrary Data Modality
- PIRF: Physics-Informed Reward Fine-Tuning for Diffusion Models
- Learning features from Newton's algorithm: a way to accelerate nonlinear parametrized PDE solvers
- THINNs: Thermodynamically Informed Neural Networks
- Flow marching for a generative PDE foundation model
- Physics-Informed Operator Learning for Hemodynamic Modeling
- Bayesian Physics Informed Neural Networks for Reliable Transformer Prognostics
- Learning Singularity-Encoded Green's Functions with Application to Iterative Methods
- ReBaNO: Reduced Basis Neural Operator Mitigating Generalization Gaps and Achieving Discretization Invariance
- Localized PCA-Net Neural Operators for Scalable Solution Reconstruction of Elliptic PDEs
- Physics-informed low-rank neural operators with application to parametric elliptic PDEs
- Fourier Neural Operators for Time-Periodic Quantum Systems: Learning Floquet Hamiltonians, Observable Dynamics, and Operator Growth
- HyPINO: Multi-Physics Neural Operators via HyperPINNs and the Method of Manufactured Solutions
- Causal Multi-fidelity Surrogate Forward and Inverse Models for ICF Implosions
- Error analysis for learning the time-stepping operator of evolutionary PDEs
- MD-PNOP: Equation-Recast Neural Operators for Minimal-Data Extrapolation and PDE Solver Acceleration
- Estimating Parameter Fields in Multi-Physics PDEs from Scarce Measurements
- Physics-Constrained Machine Learning for Chemical Engineering
- Neural Spline Operators for Risk Quantification in Stochastic Systems
- Goal-Oriented Low-Rank Tensor Decompositions for Numerical Simulation Data
- A Selective Review of Modern Stochastic Modeling: SDE/SPDE Numerics, Data-Driven Identification, and Generative Methods with Applications in Biomathematics
- LNN-PINN: A Unified Physics-Only Training Framework with Liquid Residual Blocks
- Unsupervised operator learning approach for dissipative equations via Onsager principle
- Diffeomorphic Neural Operator Learning
- Bayesian BiLO: Bilevel Local Operator Learning for Efficient Uncertainty Quantification of Bayesian PDE Inverse Problems with Low-Rank Adaptation
- An effective physics-informed neural operator framework for predicting wavefields
- Bridging ocean wave physics and deep learning: Physics-informed neural operators for nonlinear wavefield reconstruction in real-time
- Solved in Unit Domain: JacobiNet for Differentiable Coordinate-Transformed PINNs
- MASIV: Toward Material-Agnostic System Identification from Videos
- Blending data and physics for reduced-order modeling of systems with spatiotemporal chaotic dynamics
- Learning coupled Allen-Cahn and Cahn-Hilliard phase-field equations using Physics-informed neural operator(PINO)
- Scale-Consistent Learning for Partial Differential Equations
- Low-rank adaptive physics-informed HyperDeepONets for solving differential equations
- Accelerating Conjugate Gradient Solvers for Homogenization Problems with Unitary Neural Operators
- PICore: Physics-Informed Unsupervised Coreset Selection for Data Efficient Neural Operator Training
- Universal Fourier Neural Operators for periodic homogenization problems in linear elasticity
- The Fourier Spectral Transformer Networks For Efficient and Generalizable Nonlinear PDEs Prediction
- Operator-based machine learning framework for generalizable prediction of unsteady treatment dynamics in stormwater infrastructure
- Numerical and data-driven modeling of spall failure in polycrystalline ductile materials
- Single vs. Multiple Branches in DeepONet and S-DeepONet: Network Architecture Follows Coupling in Multiphysics Systems
- Physics-Informed Neural Operators for Generalizable and Label-Free Inference of Temperature-Dependent Thermoelectric Properties
- High precision PINNs in unbounded domains: application to singularity formulation in PDEs
- Maximal Update Parametrization and Zero-Shot Hyperparameter Transfer for Fourier Neural Operators
- Deep Learning-Based Prediction of High Explosive Induced Fluid Dynamics
- VideoPDE: Unified Generative PDE Solving via Video Inpainting Diffusion Models
- End-to-End Probabilistic Framework for Learning with Hard Constraints
- Principled Approaches for Extending Neural Architectures to Function Spaces for Operator Learning
- OmniFluids: Physics Pre-trained Modeling of Fluid Dynamics
- A composition of simplified physics-based model with neural operator for trajectory-level seismic response predictions of structural systems
- NOBLE -- Neural Operator with Biologically-informed Latent Embeddings to Capture Experimental Variability in Biological Neuron Models
- CoSynFlow: Conformal Symplectic Neural Flows for Cross-System Prediction of Dissipative Hamiltonian Dynamics
- PMNO: A novel physics guided multi-step neural operator predictor for partial differential equations
- Adaptive Correction for Ensuring Conservation Laws in Neural Operators
- PDE-Transformer: Efficient and Versatile Transformers for Physics Simulations
- Toward Knowledge-Guided AI for Inverse Design in Manufacturing: A Perspective on Domain, Physics, and Human-AI Synergy
- FOLIAGE: Towards Physical Intelligence World Models Via Unbounded Surface Evolution
- Neural Interpretable PDEs: Harmonizing Fourier Insights with Attention for Scalable and Interpretable Physics Discovery
- Physics-Informed Distillation of Diffusion Models for PDE-Constrained Generation
- FNOPE: Simulation-based inference on function spaces with Fourier Neural Operators
- Dual Natural Gradient Descent for Scalable Training of Physics-Informed Neural Networks
- Spectral-inspired Operator Learning with Limited Data and Unknown Physics
- FD-Bench: A Modular and Fair Benchmark for Data-driven Fluid Simulation
- Graph-Based Operator Learning from Limited Data on Irregular Domains
- Operator Learning for Schrödinger Equation: Unitarity, Error Bounds, and Time Generalization
- Guided Diffusion Sampling on Function Spaces with Applications to PDEs
- Physics-Guided Learning of Meteorological Dynamics for Weather Downscaling and Forecasting
- Physics-informed Reduced Order Modeling of Time-dependent PDEs via Differentiable Solvers
- Hamiltonian Neural PDE Solvers through Functional Approximation
- Hybrid Lagrangian-Eulerian Model for Lagrangian Fluid Simulation
- Mollifier Layers: Enabling Efficient High-Order Derivatives in Inverse PDE Learning
- Improving Data Fidelity via Diffusion Model-based Correction and Super-Resolution
- Sensitivity-Constrained Fourier Neural Operators for Forward and Inverse Problems in Parametric Differential Equations
- Bridging Model Predictive Control and Deep Learning for Scalable Reachability Analysis
- PeSANet: Physics-encoded Spectral Attention Network for Simulating PDE-Governed Complex Systems
- Neural Networks with Local Converging Inputs for Efficient Options Pricing Models
- Error-Conditioned Neural Solvers
- Kolmogorov-Arnold Reservoir Computing
- A Diagnostic Software Suite for Auditing Learned PDE Simulators
- Extending Neural Operators: Robust Handling of Functions Beyond the Training Set
- Di-BiLPS: Denoising induced Bidirectional Latent-PDE-Solver under Sparse Observations
- HyCOP: Hybrid Composition Operators for Interpretable Learning of PDEs
- Reliable and efficient steady CFD from surrogate predictions through Newton-Krylov correction
- Fourier analysis of the physics of transfer learning for data-driven subgrid-scale models of ocean turbulence
- Equi-Euler GraphNet: An Equivariant, Temporal-Dynamics Informed Graph Neural Network for Dual Force and Trajectory Prediction in Multi-Body Systems
- Equilibrium Conserving Neural Operators for Super-Resolution Learning
- Machine-learning-based simulation of turbulent flows over periodic hills using a hybrid U-Net and Fourier neural operator framework
- Enabling Automatic Differentiation with Mollified Graph Neural Operators
- Closed-Loop Neural Operator-Based Observer of Traffic Density
Related