Neural OperatorsDeepONetFNOOperator Learning

Neural Operators

A systematic map of the neural-operator family: from the DeepONet and FNO foundations through differentiable frameworks, physics-informed variants, datasets and benchmarks to real-world deployments.

Updated Nov 3, 2026 83 resources
Introduction

Why this collection

A PINN trains one network per equation to be solved. Neural operators take a different route: learning the solution operator itself. Train once, then solve a whole family of PDEs with different boundary conditions, parameters and geometries.

The payoff is direct: for a parameterised problem the response surface is trained once, and every new operating point becomes a forward pass. This is the most industrially valuable shape of surrogate modeling and reduced-order modeling.

The family has already split into several technical lines, and choosing between them requires seeing the differences clearly:

  • Branch structure — DeepONet's branch-trunk layout versus FNO's spectral layout, the latter being intrinsically mesh-independent.
  • How physics priors enter — writing conservation and dissipation into the architecture (GENERIC-FNO), or writing physical constraints into the loss.
  • Geometry handling — how an operator keeps its accuracy as geometry and mesh change.
  • Engineering reality — compute cost, memory footprint, and interfaces to existing solvers.

This collection groups resources along those four threads, from the foundational papers through to industrial deployments. If you are weighing a move from PINNs to neural operators, read the "Tutorials & Surveys" group first, then jump straight to "Method Frontiers & Variants".

Foundational Papers 2 resources

Fourier Neural Operator (Original Paper)

Foundational Paper
Introduces neural operators with integral kernels parameterized in Fourier space, enabling resolution-invariant PDE solving - a foundational and highly cited work in physics AI.
arXiv preprint (copyright held by authors) FNONeural OperatorFoundational
💬 Editor note: The FNO paper: mesh-independence comes from spectral parameterisation, which is why it remains the most widely adopted operator-learning baseline.

DeepONet (Original Paper)

Foundational Paper
Presents a branch-trunk architecture realizing a deep learning version of the universal operator approximation theorem - the first systematic architecture for learning nonlinear operators and still a mainstream baseline.
arXiv preprint (copyright held by authors) DeepONetNeural OperatorFoundational
💬 Editor note: The DeepONet paper: the branch-trunk layout opened the "learn the operator" route and, together with FNO, defines the two mainstream paradigms.

Frameworks & Implementations 8 resources

Fundamental Physics Neural Operators

Neural Network / AI for Science
A fundamental physics neural operator model project published by Delta Lab on HuggingFace. Refer to official repository documentation for specific functionality and performance details.
mit Neural OperatorsFundamental PhysicsDelta Lab

Neural Operator Predictors for Sampled Measurements

Neural Operator
A neural operator-based predictor model hosted on HuggingFace, designed for sampled measurement prediction tasks.
See LICENSE file Neural OperatorScientific Machine LearningSampled Measurements

Spherical Fourier Neural Operators

Neural Operator
A neural operator model based on spherical Fourier transforms, designed for learning partial differential equations on spherical geometries.
other Neural OperatorFourier Neural OperatorSpherical Data

DMD-Neural-Operator

Scientific Computing
A scientific computing model on HuggingFace that integrates Dynamic Mode Decomposition with Neural Operator for scientific data modeling and prediction tasks.
mit Dynamic Mode DecompositionNeural OperatorScientific Computing

POSEIDON

PDE Foundation Model / Downstream Benchmark
ETH Zurich CAM lab's PDE foundation model and downstream fine-tuning suite, using semigroup augmentation to mine abundant training pairs and cover 15 downstream tasks with mesh-invariant inference.
New No LICENSE file in the repository; the official model weights are labelled CC BY-NC 4.0 on HuggingFace and marked as research models Foundation ModelNeural OperatorDownstream Tasks

PhysicsNeMo

Solver / Training Framework
NVIDIA's physics-ML framework (formerly Modulus, renamed in 2025), a PyTorch-based library for PINN, neural operators and physics-constrained training - architecture, versions, comparisons and failure boundaries.
New Apache-2.0 PINNNeural OperatorNVIDIA
💬 Editor note: NVIDIA PhysicsNeMo turns neural operators into industrial-grade modules, with compute scheduling and deployment paths already worked out.

NeuralOperator

Neural Operator Library
Reference PyTorch library for FNO, TFNO and DeepONet: function-space contracts, parameters, getting started, failure modes and cross-resolution evaluation.
New MIT Neural OperatorFNODeepONet
💬 Editor note: The official NeuralOperator library, covering FNO, DeepONet, TNO and more — the first implementation to reach for when establishing a baseline.

NeuralOperator v2.0.0 Release

Scientific Computing
NeuralOperator 2.0.0 major release introduces Codano model, improves documentation, and enhances usability for training Neural Operators in PyTorch.
MIT License Neural OperatorDeep LearningPyTorch

Method Frontiers & Variants 57 resources

A Variational Optimal Transport Operator on Incompressible Flow

CFD/Fluid Dynamics
VIOT is a generative neural operator for incompressible density transport that predicts divergence-free velocity fields via feed-forward inference, replacing hour-scale optimization in adjoint fluid solvers and enabling rapid transport trajectory generation.
See LICENSE file Incompressible FlowNeural OperatorOptimal Transport

Adaptive Physics Transformer with Fused Global-Local Attention for Subsurface Energy Systems

Original Research Paper
ICML 2026 paper proposing APT, a geometry-/mesh-/physics-agnostic neural operator that fuses graph-based local feature extraction with global attention for long-range physical effects; achieves 4× saturation error reduction on 2D faulted CO2 storage vs prior SOTA, 19,873× speedup over GEOS and 6,800× over ECLIPSE.
arXiv preprint (copyright held by authors) Neural OperatorSubsurface EnergyCO2 Storage

AOT-POT: Adaptive Operator Transformation for Large-Scale PDE Pre-training

ScientificML
Proposes Adaptive Operator Transformation (AOT) to address structural diversity challenges in multi-PDE pre-training by adapting target solution operators rather than increasing model capacity, improving neural operator generalization across equations.
See LICENSE file Neural OperatorPDE Pre-trainingAdaptive Transformation

Fourier Neural Operators for Rayleigh-Bénard Convection

CFD
Proposes an improved Fourier Neural Operator for modeling 2D Rayleigh-Bénard convection by predicting time increments, achieving a compact model (314k parameters, 7ms inference) while revealing accuracy limitations imposed by training data resolution.
See LICENSE file Fourier Neural OperatorRayleigh-Bénard ConvectionFluid Mechanics

GENERIC-FNO: Embedding Energy Conservation and Entropy Production into Fourier Neural Operators

ScientificML
GENERIC-FNO embeds the metriplectic degeneracy structure of nonequilibrium thermodynamics into Fourier neural operators, coupling reversible energy-conserving and irreversible entropy-producing dynamics via degeneracy conditions, extending structure-preserving neural operators beyond single conservation laws.
See LICENSE file neural operatorFourier neural operatorthermodynamics
💬 Editor note: GENERIC-FNO encodes energy conservation and entropy production into the architecture — essential reading on the "physics in the structure, not the loss" route.

Geometry-Aware Anisotropic Boundary Correction for Aerodynamic Simulation

CFD
Proposes a geometry-aware anisotropic boundary correction method using neural operators as an efficient alternative to traditional CFD solvers for aerodynamic simulation, improving physical modeling near solid boundaries.
See LICENSE file aerodynamicsboundary correctionneural operator
💬 Editor note: Incremental neural operators address error accumulation in long-horizon rollouts — the key design for using operators in multi-step evolution.

How Much Memory Do We Need? Adaptive Memory Gate for Neural Operators

ScientificML
Proposes an adaptive memory gating mechanism enabling neural operators to dynamically adjust memory weights based on observation conditions, improving adaptability in solving time-dependent PDEs.
See LICENSE file Neural OperatorsAdaptive MemoryTime-dependent PDE
💬 Editor note: An adaptive memory-gating mechanism targeting drift in long-horizon prediction, with modest compute cost and a light retrofit.

Multilevel Graph Wavelet Compressed Sensing with Scale-Aware Neural Recovery

ScientificML
Proposes Graph Wavelet Compressed Sensing (GWCS) for offline compression of graph signals via sparse wavelet-domain representations, reducing data preparation costs for scientific machine learning training.
See LICENSE file graph signal processingwavelet transformcompressed sensing

NeuIDO: Neural Intrinsic Dynamics Operator for Physics-Informed 4D World Models

ScientificML
NeuIDO proposes a neural intrinsic dynamics operator that internalizes physical constraints into world models for predicting 3D object interaction trajectories, bridging the gap between physics-informed 4D generation and true world models.
See LICENSE file World ModelPhysics-Informed GenerationNeural Operator

SPUS PDE-UNet-36M-v2 (FNS-KF) PDE Solving Model

ScientificML
A 36M-parameter UNet-based PDE solver v2 by Los Alamos National Laboratory, designed for surrogate modeling of partial differential equations in scientific computing.
See LICENSE file PDE SolvingUNetNeural Operator

ATOM: A Pretrained Neural Operator for Multitask Molecular Dynamics

Original Research Paper
ICLR 2026 paper proposing ATOM, a quasi-equivariant pretrained transformer neural operator that requires no explicit molecular graph and uses temporal attention for parallel decoding of multiple future states; curates TG80 (80 compounds, 2.5M fs trajectories) and demonstrates zero-shot generalization on MD17/RMD17/MD22.
arXiv preprint (copyright held by authors) Neural OperatorMolecular DynamicsTransformer

Transolver is a Linear Transformer: Revisiting Physics-Attention through the Lens of Linear Attention

Original Research Paper
Reframes Transolver's Physics-Attention as a special case of linear attention and proposes LinearNO (Linear Attention Neural Operator); achieves SOTA on six PDE benchmarks with 40.0% fewer parameters and 36.2% less compute, plus superior performance on AirfRANS and Shape-Net Car industrial datasets.
arXiv preprint (copyright held by authors) Neural OperatorTransformerLinear Attention
💬 Editor note: Transolver rethinks physical attention, challenging the default architectural assumptions behind Transformer-based operators.

Physics-Informed Learning Annual Review 2025-2026

Annual Survey / Physics-Informed Learning
A systematic review of 2025-2026 progress in physics-informed ML - neural operators, diffusion for PDE solving, large benchmarks and industrial deployment.
Various SurveyPhysics-InformedPINN

Cluster Attention Neural Operators for Solving Parametric Partial Differential Equations

ScientificML
This paper presents cluster attention neural operators for solving parametric partial differential equations. It targets physical-field simulation by learning solution operators to accelerate parameter-space mapping, while addressing the quadratic attention complexity associated with Transformer-based neural operators.
See LICENSE file neural operatorsparametric PDEscluster attention

Does Transolver Really Need a Transformer?

CFD
This paper provides an empirical and theoretical analysis of Transolver’s physics-attention mechanism for three-dimensional fluid dynamics, examining how unstructured mesh points are assigned to slices, tokens interact through self-attention, and results are broadcast back to the points.
See LICENSE file neural operatorsphysics-attentionfluid dynamics

Energy-Efficient Operation of Neural Operators for Virtual Sensing

Thermal Management
This paper investigates shared spatial computation for energy-efficient neural-operator virtual sensing. It evaluates operating-energy reductions in a heat-exchanger service while retaining the selected checkpoint and its evaluated predictions.
See LICENSE file Neural OperatorsVirtual SensingEnergy-Efficient Computing

Enhancing Generalization in Endwall Film Cooling Prediction: Incorporating the Superposition Principle into Transformer-Based Neural Operators

CFD
This paper proposes a physics-enhanced neural operator for turbine endwall film cooling by incorporating the superposition principle into a transformer-based model. The approach targets temperature-field prediction across cooling layouts with varying numbers of film holes, focusing on improved generalization in fluid cooling analysis.
See LICENSE file Film CoolingNeural OperatorTransformer

Function-Space Transformer with Adaptive Anchors

ScientificML
A function-space Transformer with adaptive anchors targets continuous-domain physical fields, addressing the trade-off between resolving localized variation and computation over fixed uniform grids. The abstract relates this approach to neural operators and latent-attention architectures.
See LICENSE file physical fieldsneural operatorscontinuous-domain modeling

KernelOnet: An Interpretable Neural Operator Based on Kernel Functions

ScientificML
KernelOnet is an interpretable neural operator framework that incorporates kernel functions explicitly into its architecture, aligning the operator structure with boundary-type kernel-expansion forms for physics modeling.
See LICENSE file neural operatorskernel functionsinterpretable machine learning

Learning Operators of Geometry with an Interface Autoencoder

ScientificML
Proposes Interface Autoencoder (IAE) for learning neural operators on geometry-dependent PDEs, representing interface-function states in a fixed domain via finite projection codes.
See LICENSE file Neural OperatorsPDEInterface Autoencoder

MENO: Memory-Efficient Neural Operator

ScientificML
A high-performance PDE neural solver based on Manifold Function Encoder, featuring resolution-independent memory footprint and arbitrary PDE input support for large-scale physical simulations.
See LICENSE file Neural OperatorPDE SolverMemory Efficiency

Physics-Informed Neural Operator Surrogate for 2D Magnetohydrodynamic Reconnection

Electromagnetics
A physics-informed neural operator learns the map between function spaces in magnetohydrodynamic simulations, enabling efficient state inference for magnetic reconnection at any parameter and time step, significantly reducing computational cost at high Lundquist numbers.
See LICENSE file MagnetohydrodynamicsMagnetic ReconnectionNeural Operator

Preconditioned Physics-Informed Neural Operator Training

ScientificML
This paper studies preconditioned training for physics-informed neural operators across general physics simulations, using governing equations rather than solver-generated datasets to reduce offline data-generation cost and enable fresh samples during optimization.
See LICENSE file physics-informed neural operatorpreconditioned trainingscientific machine learning

Rate-Agnostic Bioacoustics: Heterogeneous Multi-Taxa Classification with Continuous Filterbanks and Fourier Neural Operators

Acoustics
This bioacoustics paper introduces a sampling-frequency-independent frontend that processes recordings at their native rates, combined with a Fourier Neural Operator backbone for heterogeneous multi-taxa classification while reducing high-frequency information loss from resampling.
See LICENSE file bioacousticsheterogeneous sampling ratesFourier Neural Operator

Scale-Split Neural Operator for Memory- and Data-Efficient 3D Turbulence Prediction

CFD
ScaleSplit-NO is a neural-operator method for three-dimensional turbulence prediction that exploits turbulence scale structure, reducing memory requirements and reliance on costly full-resolution training data in high-resolution simulation workflows.
See LICENSE file Neural Operator3D TurbulenceTurbulence Prediction

SINO: Scale-Invariant Neural Operator

ScientificML
SINO is a scale-invariant neural operator for partial differential equation solving, using low-rank parameterization to represent unresolved physics and closure mechanisms in coarse-grid computations.
See LICENSE file scale-invariant neural operatorpartial differential equationsphysics closure

Structure-preserving Fourier Neural Operators for Long-time Cahn-Hilliard Dynamics under Coarse Temporal Supervision

Materials
The paper presents a structure-preserving Fourier Neural Operator for long-time Cahn-Hilliard phase-field dynamics using coarse temporal supervision, targeting efficient surrogate prediction during computationally intensive late-stage coarsening while addressing error accumulation in repeated time stepping.
See LICENSE file Cahn-Hilliard dynamicsFourier neural operatorphase-field simulation

When Does Domain Adaptation Help on Physical Vibration Sensors? A Held-Out-Bearing Study of Neural-Operator and Convolutional Models

Structural
This paper studies rolling-element bearing fault diagnosis from vibration, a physical-sensing task, under operating-condition shift and a held-out-bearing protocol. It compares neural-operator and convolutional models while assigning each physical bearing entirely to either training or testing.
See LICENSE file Bearing Vibration DiagnosisDomain AdaptationNeural Operators

Datasets & Benchmarks 4 resources

CFDBench: Computational Fluid Dynamics Benchmark Dataset

CFD/Computational Fluid Dynamics
A large-scale benchmark dataset for machine learning in computational fluid dynamics, covering four classic CFD problems—lid-driven cavity, laminar pipe flow, dam-break, and flow around a cylinder—to evaluate neural operator generalization under unseen conditions.
apache-2.0 computational fluid dynamicsbenchmark datasetneural operators
💬 Editor note: CFDBench offers a unified multi-scenario CFD benchmark for measuring operator generalisation across flow regimes.

Cross-Representation PDE Evaluation Dataset

ScientificML
A PDE evaluation dataset containing 32 physical systems, each with four representations: source code, natural-language description, governing equation, and numerical trajectory, with both correct and corrupted variants.
See LICENSE file PDE evaluationcross-representation learningphysical systems

PDE Surrogate Benchmark Dataset

ScientificML
A benchmark dataset for PDE surrogate tasks within the ai4sci-tasks evaluation framework, containing training trajectories and evaluation splits for assessing neural operator and surrogate modeling performance.
See LICENSE file PDE SurrogateBenchmark DatasetScientific ML

PDEControl_DPC - TI-DeepONet Training Datasets

ScientificML
Simulation training datasets for PDE control containing 3000 controlled trajectories generated by classical numerical solvers under randomly sampled control sequences, designed for training TI-DeepONet surrogate models.
See LICENSE file PDE controlDeepONetNeural operator

Tutorials & Surveys 10 resources

Physics-Based Deep Learning (Online Book)

Textbook / Systematic Course
A free online textbook from TU Munich covering physics + deep learning: differentiable physics, PINNs, neural operators and turbulence modeling, with runnable code.
CC BY-NC-SA 4.0 TextbookDifferentiable PhysicsPINN
💬 Editor note: An engineering-perspective comparison: architectures benchmarked side by side on Poisson and linear-elasticity problems, plus demonstrations as Bayesian-inversion surrogates.

Operator Learning: A Survey

Survey / Reading Material
A systematic survey of operator learning covering the theory, architectures and error analysis of DeepONet, FNO and their variants - the reference entry point to learning infinite-dimensional maps.
arXiv preprint (copyright held by authors) Operator LearningSurveyDeepONet
💬 Editor note: A systematic treatment of DeepONet and FNO — theory, architecture and error analysis; the best reference for infinite-dimensional mappings.

NVIDIA PhysicsNeMo Physics-AI Framework

General SciML / Industrial Framework
NVIDIA open-source physics-AI training/inference framework (formerly Modulus) with official examples across CFD, structural, EM, weather, and semiconductor TCAD.
New Apache-2.0 PhysicsNeMoModulusIndustrial Deployment

PDEBench Scientific Machine Learning Benchmark

General SciML / Benchmark
A unified benchmark (Takamoto et al. 2022) spanning 12+ PDE families that turns "is this operator-learning model good" into comparable numbers.
New See LICENSE file BenchmarkDatasetNeural Operator

FNO (Fourier Neural Operator) Tutorial

Fourier Neural Operator / Beginner Tutorial
Systematic FNO from theory to code - for engineers and researchers wanting neural operators as PDE surrogates.
MIT FNONeural OperatorSurrogate

DeepONet Getting Started

Neural Operator / Beginner Tutorial
Systematic DeepONet theory and DeepXDE implementation - for PINN developers handling multiphysics-parameter scenarios.
MIT DeepONetNeural OperatorDeepXDE

DeepXDE Physics-Informed Neural Network Framework

General SciML / Framework Tutorial
The most mature open-source PINN/DeepONet solver framework - one API covering forward, inverse, and multiphysics problems, with Chinese-friendly docs.
New LGPL-2.1 PINNDeepONetFramework

Industrial Deployments 2 resources

Semiconductor TCAD Surrogates: FP-DDM Domain Decomposition and Etch Process Surrogates

Semiconductor / Device and Process Simulation
Semiconductor TCAD surrogates anchored on two verifiable foundations: the fp_ddm sub-example in NVIDIA PhysicsNeMo, which applies overlapping Schwarz domain decomposition to a steady-state 2D heat-conduction problem against a plane-stress elasticity case, and the etch-process morphology surrogates reported by the SK hynix TCAD Intelligence team. The entry separates process-level TCAD from device-level electrical surrogates and records the accuracy and geometry-scaling limits of each.
New Apache-2.0 for the PhysicsNeMo repository; DAC 2026 paper copyright held by the authors SemiconductorTCADDomain Decomposition

FourCastNet: Global Weather Forecasting with Adaptive Fourier Neural Operators

Weather / Data-Driven Forecasting
A global data-driven weather forecasting model proposed by NVIDIA and NERSC in 2022 that produces a week-long forecast at 0.25-degree resolution in under two seconds, matching ECMWF IFS at short lead times for large-scale variables and beating it for fine-scale fields, while making thousand-member large ensembles an inexpensive routine operation.
New Code is BSD 3-Clause (Copyright (c) 2022, FourCastNet authors); data follows the Copernicus CDS terms; model weights ship with the repository distribution notes Weather ForecastingFourier Neural OperatorDeep Learning
💬 Editor note: FourCastNet is one of the few large-scale industrial deployments of neural operators, showing feasibility at global weather-data scale.