Base Models for Parabolic Partial Differential Equations
Not on the map. Why?
Abstract (arXiv)
Parabolic partial differential equations (PDEs) appear in many disciplines to model the evolution of various mathematical objects, such as probability flows, value functions in control theory, and derivative prices in finance. It is often necessary to compute the solutions or a function of the solutions to a parametric PDE in multiple scenarios corresponding to different parameters of this PDE. This process often requires resolving the PDEs from scratch, which is time-consuming. To better employ existing simulations for the PDEs, we propose a framework for finding solutions to parabolic PDEs across different scenarios by meta-learning an underlying base distribution. We build upon this base distribution to propose a method for computing solutions to parametric PDEs under different parameter settings. Finally, we illustrate the application of the proposed methods through extensive experiments in generative modeling, stochastic control, and finance. The empirical results suggest that the proposed approach improves generalization to solving PDEs under new parameter regimes.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 24 of its 39 references that are in the map. First posted on arXiv 17 Jul 2024.
Builds on
24 of 39 references.
References · 24
- 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
- 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
- 2016Density estimation using Real NVPcited by 2,029
- 2019DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operatorscited by 1,965
- 2014NICE: Non-linear Independent Components Estimationcited by 1,052
- 2021Maximum Likelihood Training of Score-Based Diffusion Modelscited by 670
- 2021Characterizing possible failure modes in physics-informed neural networkscited by 632
- 2018FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Modelscited by 584
- 2017Solving high-dimensional partial differential equations using deep learningcited by 549
- 2021Learning the solution operator of parametric partial differential equations with physics-informed DeepOnetscited by 537
- 2022Fourier Neural Operator with Learned Deformations for PDEs on General Geometriescited by 416
12 more
- 2021Multiwavelet-based Operator Learning for Differential Equationscited by 229
- 2021A Variational Perspective on Diffusion-Based Generative Models and Score Matchingcited by 182
- 2022Fast Sampling of Diffusion Models via Operator Learningcited by 173
- 2023Tackling the Curse of Dimensionality with Physics-Informed Neural Networkscited by 78
- 2023normflows: A PyTorch Package for Normalizing Flowscited by 56
- 2021Meta-learning PINN loss functionscited by 51
- 2021Meta-Auto-Decoder for Solving Parametric Partial Differential Equationscited by 36
- 2021A novel meta-learning initialization method for physics-informed neural networkscited by 35
- 2020Numerically Solving Parametric Families of High-Dimensional Kolmogorov Partial Differential Equations via Deep Learningcited by 22
- 2022Robust SDE-Based Variational Formulations for Solving Linear PDEs via Deep Learningcited by 13
- 2022Minimax Optimal Kernel Operator Learning via Multilevel Trainingcited by 13
- 2015Learning Optimal Control via Forward and Backward Stochastic Differential Equationscited by 11
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.