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 2405.01319cs.LGcs.CE
收藏

在基于神经算子的物理仿真中强制施加局部性原理

Enforcing the Principle of Locality for Physical Simulations with Neural Operators

Carnegie Mellon University(卡内基梅隆大学)

Jiangce Chen, Wenzhuo Xu, Zeda Xu, Noelia Grande Gutiérrez, Sneha Prabha Narra, Christopher McComb

AI总结:

针对深度学习架构无法严格保证神经算子局部预测的局部依赖性、有限数据下收敛慢泛化差的问题,提出强制局部依赖的数据分解方法DDELD,实验证明其可显著加快基准模型训练收敛、降低大规模工程仿真测试误差。

摘要:

经典物理系统的含时偏微分方程(PDE)基于质量、动量和能量守恒定律建立,在科学与工程应用中无处不在。根据物理学中的局部性原理,这些PDE具有严格的局部依赖性,即某点的演化仅受其邻域影响,邻域大小由时间步长乘以系统中特征信息的传播速度决定。然而,深度学习架构无法严格保证局部依赖性——随着层数增加,其进行局部预测时的信息范围不可避免地会扩大。在训练数据有限的情况下,额外的无关信息会导致收敛缓慢、泛化能力下降。本文旨在解决这一问题,提出一种严格限制神经算子局部预测信息范围的数据分解方法,称为强制局部依赖的数据分解(DDELD)。针对多种物理现象的数值实验表明,在大规模工程仿真中,DDELD显著加快了基准模型的训练收敛速度,并降低了测试误差。

英文摘要:

Time-dependent partial differential equations (PDEs) for classic physical systems are established based on the conservation of mass, momentum, and energy, which are ubiquitous in scientific and engineering applications. These PDEs are strictly local-dependent according to the principle of locality in physics, which means that the evolution at a point is only influenced by the neighborhood around it whose size is determined by the length of timestep multiplied with the speed of characteristic information traveling in the system. However, deep learning architecture cannot strictly enforce the local-dependency as it inevitably increases the scope of information to make local predictions as the number of layers increases. Under limited training data, the extra irrelevant information results in sluggish convergence and compromised generalizability. This paper aims to solve this problem by proposing a data decomposition method to strictly limit the scope of information for neural operators making local predictions, which is called data decomposition enforcing local-dependency (DDELD). The numerical experiments over multiple physical phenomena show that DDELD significantly accelerates training convergence and reduces test errors of benchmark models on large-scale engineering simulations.

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。

↑