Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DPOT: Auto-Regressive Denoising Operator Transformer for Large-Scale PDE Pre-Training

About

Pre-training has been investigated to improve the efficiency and performance of training neural operators in data-scarce settings. However, it is largely in its infancy due to the inherent complexity and diversity, such as long trajectories, multiple scales and varying dimensions of partial differential equations (PDEs) data. In this paper, we present a new auto-regressive denoising pre-training strategy, which allows for more stable and efficient pre-training on PDE data and generalizes to various downstream tasks. Moreover, by designing a flexible and scalable model architecture based on Fourier attention, we can easily scale up the model for large-scale pre-training. We train our PDE foundation model with up to 0.5B parameters on 10+ PDE datasets with more than 100k trajectories. Extensive experiments show that we achieve SOTA on these benchmarks and validate the strong generalizability of our model to significantly enhance performance on diverse downstream PDE tasks like 3D data. Code is available at \url{https://github.com/thu-ml/DPOT}.

Zhongkai Hao, Chang Su, Songming Liu, Julius Berner, Chengyang Ying, Hang Su, Anima Anandkumar, Jian Song, Jun Zhu• 2024

Related benchmarks

TaskDatasetResultRank
Operator learningPDEBench DR
L2RE0.0103
28
Operator learningPDEBench SWE
L2 Relative Error (L2RE)0.0023
28
Dynamics downstream taskISO
NRMSE Rollout Step 16.14
28
Operator learningFNO-ν 1e-5
L2 Relative Error2.29
25
Operator learningPDEArena NS
L2RE2.94
25
Operator learningFNO-ν (1e-4)
L2RE0.0126
25
Operator learningPDEBench CNS (η=0.1, ζ=0.1)
L2 Relative Error (L2RE)0.0087
25
Operator learningPDEArena NS-cond
L2RE0.172
25
Operator learningCFDBench
L2RE0.0037
25
Operator learningPDEBench CNS (η=1, ζ=0.01)
L2RE1.46
25
Showing 10 of 121 rows
...

Other info

Follow for update