WaveBench: Benchmark Datasets for Wave Propagation PDEs

Wave-based imaging techniques play a critical role in diverse scientific, medical, and industrial endeavors, from discovering hidden structures beneath the Earth's surface to ultrasound diagnostics. They rely on accurate solutions to the forward and inverse problems for partial differential equations (PDEs) that govern wave propagation. Surrogate PDE solvers based on machine learning emerged as an effective approach to computing the solutions more efficiently than via classical numerical schemes. Our dataset, WaveBench, is a collection of benchmark datasets for wave propagation PDEs. Our code is available at https://github.com/wavebench/wavebench/.基于波动的成像技术在科学、医学及工业等诸多领域发挥着核心作用，其应用场景涵盖探测地球表层下的隐蔽构造、开展超声诊断等。此类技术依赖于对控制波动传播的偏微分方程（Partial Differential Equations，PDE）正问题与逆问题的精准求解。基于机器学习的PDE替代求解器，相较于经典数值格式，能够更高效地求得方程解，因此成为一类行之有效的求解方案。本数据集命名为WaveBench，是一套面向波动传播偏微分方程的基准数据集合集。本项目代码已公开，获取地址为https://github.com/wavebench/wavebench/。