arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.04340math.STstat.TH
收藏

关于切片逆回归的结构维度

On the Structural Dimension of Sliced Inverse Regression

Dongming Huang, Songtao Tian, Qian Lin

AI总结:

本研究通过证明高斯过程链接函数下SIR的第d个特征值随d指数衰减,以及低信号场景下中心空间估计的极小极大风险下界,解释了结构维度d>4时SIR降维效果差的长期难题。

摘要:

在本研究中,我们解决了一个长期存在的难题:当结构维度$d$(中心空间的维度)超过4时,切片逆回归(Sliced Inverse Regression, SIR)在充分降维中的表现通常较差。我们首先证明,在多指标模型$Y=f( \mathbf{P} \boldsymbol{X})+ε$中(其中$\boldsymbol{X}$是$p$维标准正态向量,$ε$是独立噪声,$\mathbf{P}$是从$\mathbb R^{p}$到$\mathbb R^{d}$的投影算子),若链接函数$f$服从高斯过程分布,则以高概率,$\mathrm{Cov}\left[\mathbb{E}(\boldsymbol{X}\mid Y)\right]$的第$d$个特征值$λ_{d}$满足$λ_{d}\leq C e^{-θd}$,其中$C$和$θ$为正的常数。随后我们聚焦于低信号场景,即$λ_{d}$可以任意小且不大于$d^{-8.1}$,并证明估计中心空间的极小极大风险的下界为$\frac{dp}{nλ_{d}}$。结合这两个结果,我们为$d$较大时SIR表现不佳这一困扰研究者近三十年的现象提供了令人信服的解释。本文开发的技术工具对于研究其他充分降维方法可能具有独立价值。

英文摘要:

In this work, we address the longstanding puzzle that Sliced Inverse Regression (SIR) often performs poorly for sufficient dimension reduction when the structural dimension $d$ (the dimension of the central space) exceeds 4. We first show that in the multiple index model $Y=f( \mathbf{P} \boldsymbol{X})+ε$ where $\boldsymbol{X}$ is a $p$-standard normal vector, $ε$ is an independent noise, and $\mathbf{P}$ is a projection operator from $\mathbb R^{p}$ to $\mathbb R^{d}$, if the link function $f$ follows the law of a Gaussian process, then with high probability, the $d$-th eigenvalue $λ_{d}$ of $\mathrm{Cov}\left[\mathbb{E}(\boldsymbol{X}\mid Y)\right]$ satisfies $λ_{d}\leq C e^{-θd}$ for some positive constants $C$ and $θ$. We then focus on the low signal regime where $λ_{d}$ can be arbitrarily small and not larger than $d^{-8.1}$, and prove that the minimax risk of estimating the central space is lower bounded by $\frac{dp}{nλ_{d}}$. Combining these two results, we provide a convincing explanation for the poor performance of SIR when $d$ is large, a phenomenon that has perplexed researchers for nearly three decades. The technical tools developed here may be of independent interest for studying other sufficient dimension reduction methods.

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。

↑