当前位置:
X-MOL 学术
›
J. Chem. Phys.
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
ISOKANN: Invariant subspaces of Koopman operators learned by a neural network.
The Journal of Chemical Physics ( IF 3.1 ) Pub Date : 2020-09-16 , DOI: 10.1063/5.0015132 Robert Julian Rabben 1 , Sourav Ray 1 , Marcus Weber 1
The Journal of Chemical Physics ( IF 3.1 ) Pub Date : 2020-09-16 , DOI: 10.1063/5.0015132 Robert Julian Rabben 1 , Sourav Ray 1 , Marcus Weber 1
Affiliation
The problem of determining the rate of rare events in dynamical systems is quite well-known but still difficult to solve. Recent attempts to overcome this problem exploit the fact that dynamic systems can be represented by a linear operator, such as the Koopman operator. Mathematically, the rare event problem comes down to the difficulty in finding invariant subspaces of these Koopman operators . In this article, we describe a method to learn basis functions of invariant subspaces using an artificial neural network.
中文翻译:
ISOKANN:由神经网络学习的Koopman算子的不变子空间。
确定动力学系统中稀有事件发生率的问题是众所周知的,但仍然很难解决。解决该问题的最新尝试利用了这样一个事实,即动态系统可以由线性算子(例如Koopman算子)表示。从数学上讲,罕见事件问题归结为寻找这些Koopman算子的不变子空间的困难。在本文中,我们描述了一种使用人工神经网络学习不变子空间基本函数的方法。
更新日期:2020-09-21
中文翻译:
ISOKANN:由神经网络学习的Koopman算子的不变子空间。
确定动力学系统中稀有事件发生率的问题是众所周知的,但仍然很难解决。解决该问题的最新尝试利用了这样一个事实,即动态系统可以由线性算子(例如Koopman算子)表示。从数学上讲,罕见事件问题归结为寻找这些Koopman算子的不变子空间的困难。在本文中,我们描述了一种使用人工神经网络学习不变子空间基本函数的方法。