Allen-Cahn Equation

该数据集是通过数值实验验证PIG方法在Allen-Cahn方程上的有效性而得出的。与JAX-PI方法相比，该实验显示PIG方法在显著提高收敛速度的同时，也保持了较高的最终准确度。实验是在一个拥有大约2万个参数的特定架构上进行的。该任务旨在解决偏微分方程（PDEs）。

This dataset is derived from numerical experiments conducted to validate the effectiveness of the PIG method on the Allen-Cahn equation. Compared with the JAX-PI method, the experiments demonstrate that the PIG method significantly improves convergence speed while maintaining high final accuracy. The experiments were carried out on a specific architecture with approximately 20,000 parameters. This task aims to solve partial differential equations (PDEs).

搜集汇总

背景与挑战

背景概述

该数据集介绍了一种名为Physics-Informed Gaussians (PIGs)的适应性网格表示方法，通过动态调整高斯参数的位置和形状来优化偏微分方程的数值近似解。实验结果表明，该方法在多个PDEs中表现优于传统方法，具有更高的准确性和灵活性。