论文标题
提高了不确定性原则
Sharpened Uncertainty Principle
论文作者
论文摘要
储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。
For any finite group $G$, any finite $G$-set $X$ and any field $F$, we consider the vector space $F^X$ of all functions from $X$ to $F$. When the group algebra $FG$ is semisimple and splitting, we find a specific basis $\widehat X$ of $F^X$, construct the Fourier transform: $F^X\to F^{\widehat X}$, $f\mapsto\widehat f$, and define the rank support $\mbox{rk-supp}(\widehat f)$; we prove that $\mbox{rk-supp}(\widehat f)=\dim FGf$, where $FGf$ is the submodule of the permutation module $FX$ generated by the element $f=\sum_{x\in X}f(x)x$. Next, we extend a sharpened uncertainty principle for abelian finite groups by Feng, Hollmann, and Xiang [9] to the following extensive framework: for any field $F$, any transitive $G$-set $X$ and $0\neq f\in F^X$ we prove that $$ |{\rm supp}(f)|\cdot \dim FGf \geq |X|+|{\rm supp}(f)|-|X_{{\rm supp}(f)}|, $$ where ${\rm supp}(f)$ is the support of $f$, and $X_{{\rm supp}(f)}$ is a block of $X$ associated with the subset ${\rm supp}(f)$ such that ${\rm supp}(f)$ is a disjoint union of some translations of the block. Then many (sharpened or classical) versions of finite-dimensional uncertainty principle are derived as corollaries.