Solving partial differential equations using large-data models: a literature review

Abdul Mueed Hafiz,Irfan Faiq,M. Hassaballah

DOI: https://doi.org/10.1007/s10462-024-10784-5

IF: 9.588

2024-05-26

Artificial Intelligence Review

Abstract:Mathematics lies at the heart of engineering science and is very important for capturing and modeling of diverse processes. These processes may be naturally-occurring or man-made. One important engineering problem in this regard is the modeling of advanced mathematical problems and their analysis. Partial differential equations (PDEs) are important and useful tools to this end. However, solving complex PDEs for advanced problems requires extensive computational resources and complex techniques. Neural networks provide a way to solve complex PDEs reliably. In this regard, large-data models are new generation of techniques, which have large dependency capturing capabilities. Hence, they can richly model and accurately solve such complex PDEs. Some common large-data models include Convolutional neural networks (CNNs) and their derivatives, transformers, etc. In this literature survey, the mathematical background is introduced. A gentle introduction to the area of solving PDEs using large-data models is given. Various state-of-the-art large-data models for solving PDEs are discussed. Also, the major issues and future scope of the area are identified. Through this literature survey, it is hoped that readers will gain an insight into the area of solving PDEs using large-data models and pursue future research in this interesting area.

computer science, artificial intelligence

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve

The main purpose of this paper is to explore how to use big data models (such as Convolutional Neural Networks, Recurrent Neural Networks, Long Short-Term Memory Networks, Generative Adversarial Networks, and Transformers) to solve Partial Differential Equations (PDEs). Specifically, the paper attempts to address the following issues:

1. **Solving Complex PDEs**:
- Complex partial differential equations require substantial computational resources and sophisticated techniques to solve, while neural networks provide a reliable method for solving them.
- Big data models have strong dependency-capturing capabilities, enabling them to richly model and accurately solve complex PDEs.

2. **Limitations of Existing Methods**:
- Traditional methods like the Finite Element Method (FEM) can solve high-order PDEs but require significant computational resources, and multiple iterative solutions limit their practicality.
- Other traditional methods (such as the Finite Volume Method, Particle Methods, etc.) also have their own limitations.

3. **Advantages and Disadvantages of Big Data Models**:
- Big data models like Convolutional Neural Networks (CNNs) have advantages in solving PDEs, such as simple implementation, ease of using large datasets, and consistency in parameter space.
- However, they also have drawbacks, such as the lack of theoretical convergence guarantees, longer runtime for each forward solution, and weaker theoretical analysis foundations.

Through a literature review, the paper hopes readers can gain a deep understanding of the research field of using big data models to solve PDEs and encourages them to conduct further research in this interesting area. Additionally, the paper outlines current trends, related technologies, and future research directions.