arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.06757math.FA
收藏

基于分数阶傅里叶变换的深度卷积神经网络与数据逼近

Deep convolutional neural networks and data approximation using the fractional Fourier transform

M. H. A. Biswas, P. Massopust, R. Ramakrishnan

AI总结:

该研究提出结合分数阶傅里叶变换(FrFT)的深度卷积神经网络并分析其θ-平移不变性,同时基于FrFT求解数据逼近的最优θ-平移不变空间,证明FrFT域存在对应最接近带限函数空间。

摘要:

本文第一部分定义了一种结合分数阶傅里叶变换(FrFT)的深度卷积神经网络,该网络借助与FrFT相关的θ-平移算子构建。随后研究了该网络的θ-平移不变性,与经典情形不同,这类网络不具备平移不变性。\n第二部分研究了基于FrFT的数据逼近问题:给定数据集$\fl=\f_1,\f_m\backslash\backslash\backslashsubset L^2(\R^n)$,求解得到$Φ=\backslash\phi_1,\backslash\phi_\backslash\ell\backslash\backslash\backslash$,使得$V_θ(Φ)=\backslash\argmin\sum_{j=1}^m \\|f_j-P_{V}f_j\\|^2$,其中最小值在所有由至多$\backslash\ell$个元素生成的θ-平移不变空间中取到。此外,本文证明了FrFT域中存在一个在上述意义下与$\fl$“最接近”的带限函数空间。

英文摘要:

In the first part of this paper, we define a deep convolutional neural network connected with the fractional Fourier transform (FrFT) using the $θ$-translation operator, the translation operator associated with the FrFT. Subsequently, we study $θ$-translation invariance properties of this network. Unlike the classical case, these networks are not translation invariant. \par In the second part, we study data approximation problems using the FrFT. More precisely, given a data set $\fl=\{f_1,\cdots, f_m\}\subset L^2(\R^n)$, we obtain $Φ=\{ϕ_1,\cdots,ϕ_\ell\}$ such that \[ V_θ(Φ)=\argmin\sum_{j=1}^m \|f_j-P_{V}f_j\|^2, \] where the minimum is taken over all $θ$-shift invariant spaces generated by at most $\ell$ elements. Moreover, we prove the existence of a space of bandlimited functions in the FrFT domain which is ``closest" to $\fl$ in the above sense.

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。

↑