跳到正文
返回论文列表
stat.ML提交于 已译

低秩矩阵推断中的稀疏噪声污染:PCA 基准

Sparse corruption in low-rank matrix inference: the PCA benchmark

Urte Adomaityte · Gabriele Sicuro · Pierpaolo Vivo

中文摘要

主成分分析(PCA)是从含噪观测中提取低秩信号的标准工具。已知当对受稠密、均匀噪声污染的一阶信号应用 PCA 时,在大矩阵尺寸极限下会出现 BBP 相变,即离群特征值的出现与对应特征向量的对齐在同一临界信号强度处发生。这里研究稀疏噪声污染的情形。噪声矩阵建模为具有有限平均连通度的加权无向图的邻接矩阵。利用复制方法(replica method),通过由群体动力学(population dynamics)求解的递归分布方程,解析计算了典型最大特征值、最大特征向量分量密度以及与信号的重叠。识别出两个作为图连通度函数的信号强度阈值:θ_crit 标志通过最大特征向量实现信号恢复,推广了 BBP 相变;θ_b 处信号相关特征值从体部分离。对于非零均值噪声,由于结构稀疏图离群值,这两个阈值不必重合,导致与最大特征向量的重叠出现不连续相变。对于随机正则度分布,得到两个阈值、信号特征值和平方重叠的精确闭式表达式,并证明相同公式也能预测中间区域 θ_b<θ≤θ_crit 中信号与第二特征向量的重叠。对于泊松度分布,递归的近似闭合在恢复阶段给出快速准确的预测,与群体动力学的精确结果一致。还恢复了大连通度极限下的稠密噪声结果,并通过大矩阵数值对角化验证了理论。

关键要点

  1. 01研究稀疏噪声污染下低秩矩阵推断中 PCA 的相变行为,噪声建模为有限平均连通度的加权图邻接矩阵。
  2. 02用复制方法和群体动力学求解递归分布方程,解析得到最大特征值、特征向量密度及与信号的重叠。
  3. 03识别出两个阈值 θ_crit 和 θ_b,非零均值噪声下二者不重合,导致重叠出现不连续相变。
  4. 04随机正则度分布下给出阈值、信号特征值和平方重叠的精确闭式解,并预测第二特征向量重叠。
  5. 05泊松度分布下近似闭合递归给出快速准确预测,大连通度极限恢复稠密噪声结果,数值对角化验证理论。

解读

尚无解读。

原始英文摘要

arXiv:2511.11927v3 Announce Type: replace Abstract: Principal Component Analysis (PCA) is a standard tool for extracting a low-rank signal from noisy observations. It is known that when applying PCA to a rank-one signal corrupted by a dense, homogeneous noise, in the large matrix size limit, the BBP transition occurs, where the emergence of an outlying eigenvalue and the alignment of the corresponding eigenvector occur at the same critical signal strength. Here we study the case of sparse noise corruption. The noise matrix is modelled as the adjacency matrix of a weighted undirected graph with finite average connectivity. Using the replica method, we analytically compute the typical top eigenvalue, the top eigenvector component density, and the overlap with the signal, through recursive distributional equations solved by population dynamics. We identify two signal-strength transitions as functions of graph connectivity: $\theta_{\rm crit}$, marking signal recovery by the top eigenvector and generalising the BBP transition, and $\theta_{\rm b}$, where the signal-related eigenvalue detaches from the bulk. For noise with nonzero mean, these transitions need not coincide because of a structural sparse-graph outlier, leading to a discontinuous transition in the overlap with the top eigenvector. For a Random Regular degree distribution, we obtain exact closed-form expressions for both thresholds, the signal eigenvalue and the squared overlap, and show that the same formulas also predict the overlap of the signal with the second eigenvector in the intermediate regime $\theta_{\rm b}<\theta\leq \theta_{\rm crit}$. For a Poissonian degree distribution, an approximate closure of the recursions yields fast and accurate predictions in the recovery phase that agree with exact results from population dynamics. We also recover dense-noise results in the large-connectivity limit, and validate the theory by numerical diagonalisation of large matrices.

同方向论文 · stat.ML

查看全部 →