Abstract
Significance Partial differential equations (PDEs) are among the most ubiquitous tools used in modeling problems in nature. However, solving high-dimensional PDEs has been notoriously difficult due to the “curse of dimensionality.” This paper introduces a practical algorithm for solving nonlinear PDEs in very high (hundreds and potentially thousands of) dimensions. Numerical results suggest that the proposed algorithm is quite effective for a wide variety of problems, in terms of both accuracy and speed. We believe that this opens up a host of possibilities in economics, finance, operational research, and physics, by considering all participating agents, assets, resources, or particles together at the same time, instead of making ad hoc assumptions on their interrelationships.
Keywords
Partial differential equationDeep learningComputer scienceApplied mathematicsMathematicsMathematical analysisArtificial intelligence
Affiliated Institutions
Related Publications
Scientific Machine Learning Through Physics–Informed Neural Networks: Where we are and What’s Next
Abstract Physics-Informed Neural Networks (PINN) are neural networks (NNs) that encode model equations, like Partial Differential Equations (PDE), as a component of the neural n...
Deep Reinforcement Learning for Multiagent Systems: A Review of Challenges, Solutions, and Applications
Reinforcement learning (RL) algorithms have been around for decades and employed to solve various sequential decision-making problems. These algorithms, however, have faced grea...
A Global Geometric Framework for Nonlinear Dimensionality Reduction
Scientists working with large volumes of high-dimensional data, such as global climate patterns, stellar spectra, or human gene distributions, regularly confront the problem of ...
Scaling learning algorithms towards AI
One long-term goal of machine learning research is to produce methods that are applicable to highly complex tasks, such as perception (vision, audition), rea-soning, intelligent...
Reducing the Dimensionality of Data with Neural Networks
High-dimensional data can be converted to low-dimensional codes by training a multilayer neural network with a small central layer to reconstruct high-dimensional input vectors....
Publication Info
- Year
- 2018
- Type
- article
- Volume
- 115
- Issue
- 34
- Pages
- 8505-8510
- Citations
- 1560
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
1560
OpenAlex
113
Influential
Cite This
Jiequn Han,
Arnulf Jentzen,
E Weinan
(2018).
Solving high-dimensional partial differential equations using deep learning.
Proceedings of the National Academy of Sciences
, 115
(34)
, 8505-8510.
https://doi.org/10.1073/pnas.1718942115
Identifiers
- DOI
- 10.1073/pnas.1718942115
- PMID
- 30082389
- PMCID
- PMC6112690
- arXiv
- 1707.02568