TT-PINN: A Tensor-Compressed Neural PDE Solver for Edge Computing
TT-PINN: A Tensor-Compressed Neural PDE Solver for Edge Computing
复制标题
DOI：
10.48550/arxiv.2207.01751
复制
发表时间：
2022-07
期刊：
影响因子：
--
通讯作者：
Z. Liu;Xinling Yu;Zheng Zhang
中科院分区：
文献类型：
--
作者：
Z. Liu;Xinling Yu;Zheng Zhang
物理信息神经网络（pinn）由于其对复杂物理系统建模的能力而得到越来越多的应用。为了获得更好的表达能力，在许多问题中需要越来越大的网络规模。当我们需要在内存、计算和能源有限的边缘设备上训练pin码时，这就带来了挑战。为了能够在边缘设备上训练pin网络，本文提出了一种基于张量-训练分解的端到端压缩pin网络。在求解亥姆霍兹方程时，我们提出的模型明显优于具有较少参数的原始pinn，并且在总体参数减少高达15美元的情况下获得令人满意的预测。
Physics-informed neural networks (PINNs) have been increasingly employed due to their capability of modeling complex physics systems. To achieve better expressiveness, increasingly larger network sizes are required in many problems. This has caused challenges when we need to train PINNs on edge devices with limited memory, computing and energy resources. To enable training PINNs on edge devices, this paper proposes an end-to-end compressed PINN based on Tensor-Train decomposition. In solving a Helmholtz equation, our proposed model significantly outperforms the original PINNs with few parameters and achieves satisfactory prediction with up to 15$\times$ overall parameter reduction.