Understanding the Efficacy of U-Net & Vision Transformer for Groundwater Numerical Modelling
Abstract
This paper presents a comprehensive comparison of various machine learning models, namely U-Net, U-Net integrated with Vision Transformers (ViT), and Fourier Neural Operator (FNO), for time-dependent forward modelling in groundwater systems. Through testing on synthetic datasets, it is demonstrated that U-Net and U-Net + ViT models outperform FNO in accuracy and efficiency, especially in sparse data scenarios. These findings underscore the potential of U-Net-based models for groundwater modelling in real-world applications where data scarcity is prevalent.
Cite
Text
Taccari et al. "Understanding the Efficacy of U-Net & Vision Transformer for Groundwater Numerical Modelling." ICML 2023 Workshops: SynS_and_ML, 2023.Markdown
[Taccari et al. "Understanding the Efficacy of U-Net & Vision Transformer for Groundwater Numerical Modelling." ICML 2023 Workshops: SynS_and_ML, 2023.](https://mlanthology.org/icmlw/2023/taccari2023icmlw-understanding/)BibTeX
@inproceedings{taccari2023icmlw-understanding,
title = {{Understanding the Efficacy of U-Net & Vision Transformer for Groundwater Numerical Modelling}},
author = {Taccari, Maria Luisa and Ovadia, Oded and Wang, He and Chen, Xiaohui and Kahana, Adar and Jimack, Peter},
booktitle = {ICML 2023 Workshops: SynS_and_ML},
year = {2023},
url = {https://mlanthology.org/icmlw/2023/taccari2023icmlw-understanding/}
}