arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2205.06308math.PRstat.ML
收藏

带多项式标度的随机内积核矩阵谱的等价原理

An Equivalence Principle for the Spectrum of Random Inner-Product Kernel Matrices with Polynomial Scalings

Yue M. Lu, Horng-Tzer Yau

AI总结:

该研究在多项式标度机制下推导了随机内积核矩阵经验谱分布的弱极限,提出其谱与Wishart矩阵和高斯正交系综矩阵线性组合的谱渐近等价,推广了线性标度下的已有结果,并拓展至迷向高斯数据场景。

摘要:

我们研究了一类随机矩阵,其元素通过对从ℝ^d中单位球面上均匀抽取的n个独立数据向量之间的两两内积应用非线性核函数得到。本研究的动机来自机器学习与统计学领域的应用,这类核随机矩阵及其谱性质在其中发挥着重要作用。我们在多项式标度机制下建立了这类矩阵经验谱分布的弱极限,该机制下d、n趋于无穷大,满足n/d^ℓ→κ,其中ℓ为固定自然数,κ取值于(0, ∞)。我们的结果推广了Cheng和Singer此前的一项结论,他们在线性标度机制(即ℓ=1)下研究了同一模型。\n 我们的工作揭示了一条等价原理:随机核矩阵的谱与一个更简单的矩阵模型的谱渐近等价,该简单模型由一个(带平移的)Wishart矩阵和一个从高斯正交系综中采样的独立矩阵线性组合构造而成。Wishart矩阵的纵横比以及线性组合的系数由ℓ和核函数在正交Hermite多项式基下的展开式决定。由此可得,随机核矩阵的极限谱可以表征为Marchenko-Pastur律与半圆律之间的自由加性卷积。我们还将结果推广到了数据向量采样自迷向高斯分布而非球面分布的情形。

英文摘要:

We investigate random matrices whose entries are obtained by applying a nonlinear kernel function to pairwise inner products between $n$ independent data vectors, drawn uniformly from the unit sphere in $\mathbb{R}^d$. This study is motivated by applications in machine learning and statistics, where these kernel random matrices and their spectral properties play significant roles. We establish the weak limit of the empirical spectral distribution of these matrices in a polynomial scaling regime, where $d, n \to \infty$ such that $n / d^\ell \to κ$, for some fixed $\ell \in \mathbb{N}$ and $κ\in (0, \infty)$. Our findings generalize an earlier result by Cheng and Singer, who examined the same model in the linear scaling regime (with $\ell = 1$). Our work reveals an equivalence principle: the spectrum of the random kernel matrix is asymptotically equivalent to that of a simpler matrix model, constructed as a linear combination of a (shifted) Wishart matrix and an independent matrix sampled from the Gaussian orthogonal ensemble. The aspect ratio of the Wishart matrix and the coefficients of the linear combination are determined by $\ell$ and the expansion of the kernel function in the orthogonal Hermite polynomial basis. Consequently, the limiting spectrum of the random kernel matrix can be characterized as the free additive convolution between a Marchenko-Pastur law and a semicircle law. We also extend our results to cases with data vectors sampled from isotropic Gaussian distributions instead of spherical distributions.

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。

↑