Research map / Probabilistic and scientific modeling
Differential equations and PDEs: research map
676 papers on Differential equations and PDEs in Probabilistic and scientific modeling, from ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026), grouped into 2 clusters and 7 approaches. The busiest year so far is 2026.
What it is
Differential equations describe weather, fluids, materials and much more, but solving them numerically is slow. This topic trains neural networks to solve or approximate them much faster (neural PDE solvers and neural operators) and uses differential equations inside networks (neural ODEs).
Terms you will see on this page:
- PDE
- partial differential equation, relating rates of change in space and time.
- Neural operator
- a network that learns a map between whole functions, such as initial condition to solution.
- Physics-informed neural network (PINN)
- a network trained so its output satisfies the equation.
Start reading with the most cited papers here:
- Fourier Neural Operator for Parametric Partial Differential Equations (ICLR 2021 · 1,068 citations)
- Neural Ordinary Differential Equations (NeurIPS 2018 · 558 citations)
Within Probabilistic and scientific modeling, its share grew from 8.8% in 2023–24 to 9.8% in 2025–26 (178 → 299 papers at ICML, NeurIPS, ICLR, CVPR and AAAI, the venues with data for all four years).
Explore Differential equations and PDEs in the interactive map
Working on something in this topic? Describe your idea in scime atlas to see which approach it falls under, the closest papers by meaning and how crowded the spot has become.
What changed recently
Approaches whose share of the topic moved most between 2023–24 and 2025–26 (ICML, NeurIPS, ICLR, CVPR and AAAI):
- Physics-informed operator learning rising: 18% → 26% of the topic (32 → 78 papers)
- Physics-informed neural network training rising: 11% → 15% of the topic (20 → 44 papers)
- Neural operators for general geometries rising: 20% → 22% of the topic (35 → 66 papers)
- Neural PDE solvers falling: 26% → 21% of the topic (47 → 64 papers)
- Neural ODE training and analysis falling: 11% → 8% of the topic (20 → 23 papers)
Papers from 2025–2026 that are already cited most:
- Neural Variable-Order Fractional Differential Equation Networks (AAAI 2025 · 9 citations)
Approaches and key papers
Neural PDE solvers · 495 papers
Keywords: pde, physics, partial differential
Approaches in this cluster:
- Neural PDE solvers (146 papers)
Solves PDEs with convergence guarantees, coarse correction and latent spectral models. - Physics-informed operator learning (149 papers)
Embeds physical constraints in DeepONets and diffusion models for fluid and graph PDEs. - Neural operators for general geometries (123 papers)
Learns PDE solution operators on irregular domains using neural fields and latent operators. - Physics-informed neural network training (77 papers)
Improves PINN training through sequential, accelerated and competitive learning.
Most cited and most cited since 2024:
- Fourier Neural Operator for Parametric Partial Differential Equations (ICLR 2021 · 1,068 citations)
- Implicit Neural Representations with Periodic Activation Functions (NeurIPS 2020 · 262 citations)
- Fourier Neural Operator with Learned Deformations for PDEs on General Geometries (NeurIPS 2024 · 150 citations)
- Challenges in Training PINNs: A Loss Landscape Perspective (ICML 2024 · 40 citations)
Neural ODEs · 181 papers
Keywords: ordinary, odes, differential equations
Approaches in this cluster:
- Neural ODE training and analysis (83 papers)
Study and speed up neural ODEs through adaptive control, theory, and simulation-free training. - Probabilistic inference for ODEs and SDEs (65 papers)
Estimate parameters and solutions of differential equations with probabilistic solvers and state-space inference. - Structured continuous-depth models (33 papers)
Build continuous-depth and differential-equation-based networks with conservation, sparsity, or higher-order structure.
Most cited and most cited since 2024:
- Neural Ordinary Differential Equations (NeurIPS 2018 · 558 citations)
- DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps (NeurIPS 2022 · 284 citations)
- Signed Graph Neural Ordinary Differential Equation for Modeling Continuous-Time Dynamics (AAAI 2024 · 14 citations)
- Neural Variable-Order Fractional Differential Equation Networks (AAAI 2025 · 9 citations)
Related topics in Probabilistic and scientific modeling
- Dynamical systems (1,019)
- Uncertainty and Gaussian processes (1,421)
- Molecules and proteins (2,120)
- Variational and generative sampling (2,370)
- Treatment effects (468)
- Causal discovery (973)
- Bayesian optimization (434)
