arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.10013cs.DC
收藏

基于区域分解的并行物理信息神经网络

Parallel Physics-Informed Neural Networks via Domain Decomposition

Khemraj Shukla, Ameya D. Jagtap, George Em Karniadakis

摘要:

我们开发了一个分布式框架的物理信息神经网络(PINNs)的基础上,最近的两个扩展,即保守PINNs(cPINNs)和扩展PINNs(XPINNs),分别采用域分解在空间和时空。这种区域分解赋予cPINN和XPINN与vanilla PINN相比的几个优点,例如并行化能力,大表示能力,高效的超参数调整,并且对于多尺度和多物理问题特别有效。在这里,我们提出了一个并行算法的cPINN和XPINN构造与MPI $+$ X,其中X $\in \{\text{CPUs},~\text{GPUs}\}$描述的混合编程模型。cPINN和XPINN相对于更经典的数据和模型并行方法的主要优点是在每个子域中分别优化每个神经网络的所有超参数的灵活性。我们比较了分布式cPINN和XPINN的性能,各种前向问题,使用弱和强缩放。我们的研究结果表明,对于空间域分解,cPINN在通信成本方面更有效,但XPINN提供了更大的灵活性,因为它们还可以处理任何微分方程的时域分解,并且可以处理任何任意形状的复杂子域。为此,我们还提出了一个应用程序的并行XPINN方法求解逆扩散问题的可变电导率在美国地图上,使用10个区域作为子域。

英文摘要:

We develop a distributed framework for the physics-informed neural networks (PINNs) based on two recent extensions, namely conservative PINNs (cPINNs) and extended PINNs (XPINNs), which employ domain decomposition in space and in time-space, respectively. This domain decomposition endows cPINNs and XPINNs with several advantages over the vanilla PINNs, such as parallelization capacity, large representation capacity, efficient hyperparameter tuning, and is particularly effective for multi-scale and multi-physics problems. Here, we present a parallel algorithm for cPINNs and XPINNs constructed with a hybrid programming model described by MPI $+$ X, where X $\in \{\text{CPUs},~\text{GPUs}\}$. The main advantage of cPINN and XPINN over the more classical data and model parallel approaches is the flexibility of optimizing all hyperparameters of each neural network separately in each subdomain. We compare the performance of distributed cPINNs and XPINNs for various forward problems, using both weak and strong scalings. Our results indicate that for space domain decomposition, cPINNs are more efficient in terms of communication cost but XPINNs provide greater flexibility as they can also handle time-domain decomposition for any differential equations, and can deal with any arbitrarily shaped complex subdomains. To this end, we also present an application of the parallel XPINN method for solving an inverse diffusion problem with variable conductivity on the United States map, using ten regions as subdomains.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑