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 2402.05563math.NAcs.AIcs.NA
收藏

神经多重网格架构

Neural Multigrid Architectures

Skolkovo Institute of Science and Technology(斯科尔科沃科技学院)

Vladimir Fanaskov

AI总结:

提出一种简洁的无矩阵神经多重网格架构,通过参数共享和层序列化训练,在数千至数百万未知数问题上高效,可为几何多重网格找到最优光滑器,使误差传播矩阵谱半径缩小2-5倍。

摘要:

我们提出了一种适用于多重网格方法的便捷无矩阵神经架构。该架构足够简单,可以用不到五十行代码实现,但它涵盖了大量不同的多重网格求解器。我们认为,一个没有密集层的固定神经网络无法实现高效的迭代方法。因此,标准训练协议无法产生具有竞争力的求解器。为了克服这一困难,我们采用了参数共享和层的序列化。由此产生的网络可以在具有数千个未知数的线性问题上进行训练,并在具有数百万个未知数的问题上保持其效率。从数值线性代数的角度来看,网络的训练对应于为几何多重网格方法寻找最优光滑器。我们在几个二阶椭圆方程上演示了我们的方法。对于测试的线性系统,与使用Jacobi光滑器的基本线性多重网格相比,我们获得了误差传播矩阵的谱半径缩小两到五倍的结果。

英文摘要:

We propose a convenient matrix-free neural architecture for the multigrid method. The architecture is simple enough to be implemented in less than fifty lines of code, yet it encompasses a large number of distinct multigrid solvers. We argue that a fixed neural network without dense layers can not realize an efficient iterative method. Because of that, standard training protocols do not lead to competitive solvers. To overcome this difficulty, we use parameter sharing and serialization of layers. The resulting network can be trained on linear problems with thousands of unknowns and retains its efficiency on problems with millions of unknowns. From the point of view of numerical linear algebra network's training corresponds to finding optimal smoothers for the geometric multigrid method. We demonstrate our approach on a few second-order elliptic equations. For tested linear systems, we obtain from two to five times smaller spectral radius of the error propagation matrix compare to a basic linear multigrid with Jacobi smoother.

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。

↑