Go to DMLR homepageSuperBench: A Super-Resolution Benchmark Dataset for Scientific Machine Learning
Abstract: Super-resolution (SR) techniques aim to enhance data resolution, enabling the retrieval of finer details, and improving the overall quality and fidelity of the data representation. There is growing interest in applying SR methods to complex spatiotemporal systems within the Scientific Machine Learning (SciML) community, with the hope of accelerating numerical simulations and/or improving forecasts in weather, climate, and related areas. However, the lack of standardized benchmark datasets for comparing and validating SR methods hinders progress and adoption in SciML. To address this, we introduce SuperBench (https://github.com/erichson/SuperBench), the first benchmark dataset featuring high-resolution datasets (up to $2048\times2048$ dimensions), including data from fluid flows, cosmology, and weather. Here, we focus on validating spatial SR performance from data-centric and physics-preserved perspectives, as well as assessing robustness to data degradation tasks. While deep learning-based SR methods (developed in the computer vision community) excel on certain tasks, despite relatively limited prior physics information, we identify limitations of these methods in accurately capturing intricate fine-scale features and preserving fundamental physical properties and constraints in scientific data. These shortcomings highlight the importance and subtlety of incorporating domain knowledge into ML models. We anticipate that SuperBench will help to advance SR methods for science.
Certifications: Dataset Certification, Reproducibility Certification
Keywords: super-resolution, scientific machine learning, benchmark dataset
Changes Since Last Submission: In response to the reviewers' comments, we have made several small revisions to the manuscript:
* We clarified the reasoning behind our focus on 2D super-resolution, explained the limitations of using 2D slices from 3D simulations.
* We refined our discussion of the computational requirements.
* We addressed concerns about the responsible use of our benchmark by emphasizing the importance of rigorous validation to prevent misleading results in critical scientific applications (see App. G).
* We added additional results for SRGAN.
Video: https://www.youtube.com/watch?v=4f64LwtNMBY&t=55s
Code: https://github.com/erichson/SuperBench
Assigned Action Editor: ~Holger_Caesar2
Submission Number: 73
Loading