High dimensional statistics

March 12, 2015 — October 28, 2021

functional analysis
high d
linear algebra
probability
signal processing
sparser than thou

Placeholder to think about the many weird problems arising in very high dimensional statistical inference. There are many approaches to this problem: throwing out dimensions/predictors as in model selection, considering low dimensional projections, viewing objects with matrix structure for concentration or factorisation, or tensor structure even.

1 Soap bubbles

High dimensional distributions are extremely odd, and concentrate in weird ways. For example, for some natural definitions of typical, typical items are not average items in high See Sander Dielemann’s musings on typicality for an introduction to this plus some motivating examples.

For another example, consider this summary result of Vershynin (2015):

Let \(K\) be an isotropic convex body (e.g. an \(L_2\) ball) in \(\mathbb{R}^{n},\) and let \(X\) be a random vector uniformly distributed in \(K\), with \(\mathbb{E}X=0\) and \(\mathbb{E}XX^{\top}=I_n.\) Then the following is true for some positive constants \(C,c\):

  1. (Concentration of volume) For every \(t \geq 1\), one has \[ \mathbb{P}\left\{\|X\|_{2}>t \sqrt{n}\right\} \leq \exp (-c t \sqrt{n}) \]
  2. (Thin shell) For every \(\varepsilon \in(0,1),\) one has \[ \mathbb{P}\left\{\left|\|X\|_{2}-\sqrt{n}\right|>\varepsilon \sqrt{n}\right\} \leq C \exp \left(-c \varepsilon^{3} n^{1 / 2}\right) \]

That is, even with the mass uniformly distributed over space, as the dimension grows, it all ends up in a thin shell, because volume grows exponentially in dimension. This is popularly known as a soap bubble phenomenon. This is one of the phenomena that leads to interesting behaviour in low dimensional projection. The more formal name is the Gaussian Annulus Theorem. Turning it around, for a d-dimensional spherical Gaussian with unit variance in each direction, for any \(\beta \leq \sqrt{d}\), all but at most \(3 e^{-c \beta^{2}}\) of the probability mass lies within the annulus \(\sqrt{d}-\beta \leq|\mathbf{x}| \leq \sqrt{d}+\beta,\) where \(c\) is a fixed positive constant.

Figure 1

2 Convex hulls

Balestriero, Pesenti, and LeCun (2021) cite Bárány and Füredi (1988):

Given a \(d\)-dimensional dataset \(\boldsymbol{X} \triangleq\left\{\boldsymbol{x}_{1}, \ldots, \boldsymbol{x}_{N}\right\}\) with i.i.d. samples \(\boldsymbol{x}_{n} \sim \mathcal{N}\left(0, I_{d}\right), \forall n\), the probability that a new sample \(\boldsymbol{x} \sim \mathcal{N}\left(0, I_{d}\right)\) is in interpolation regime (recall Def. 1 ) has the following limiting behavior \[ \lim _{d \rightarrow \infty} p(\underbrace{\boldsymbol{x} \in \operatorname{Hull}(\boldsymbol{X})}_{\text {interpolation }})= \begin{cases}1 & \Longleftrightarrow N>d^{-1} 2^{d / 2} \\ 0 & \Longleftrightarrow N<d^{-1} 2^{d / 2}\end{cases} \]

They observe that this implies high dimensional statistics rarely interpolates between data points, which is not surprising, but only in retrospect. Despite some expertise in high-dimensional problems I had never noticed this fact myself. Interestingly they collect evidence that suggests that low-d projections and latent spaces are also rarely interpolating.

3 Empirical processes in high dimensions

Combining empirical process theory with high dimensional statistics gets us to some interesting models. See, e.g. van de Geer (2014b).

4 Markov Chain Monte Carlo in high dimensions

TBD

5 References

Balestriero, Pesenti, and LeCun. 2021. Learning in High Dimension Always Amounts to Extrapolation.” arXiv:2110.09485 [Cs].
Bárány, and Füredi. 1988. On the Shape of the Convex Hull of Random Points.” Probability Theory and Related Fields.
Borgs, Chayes, Cohn, et al. 2014. An \(L^p\) Theory of Sparse Graph Convergence I: Limits, Sparse Random Graph Models, and Power Law Distributions.” arXiv:1401.2906 [Math].
Bühlmann, and van de Geer. 2011. Statistics for High-Dimensional Data: Methods, Theory and Applications.
———. 2015. High-Dimensional Inference in Misspecified Linear Models.” arXiv:1503.06426 [Stat].
Candès, Romberg, and Tao. 2006. Robust Uncertainty Principles: Exact Signal Reconstruction from Highly Incomplete Frequency Information.” IEEE Transactions on Information Theory.
Chen, and Wang. n.d. Discussion on ‘Confidence Intervals and Hypothesis Testing for High-Dimensional Regression’.”
Chernozhukov, Chetverikov, Demirer, et al. 2018. Double/Debiased Machine Learning for Treatment and Structural Parameters.” The Econometrics Journal.
Chernozhukov, Hansen, Liao, et al. 2018. Inference For Heterogeneous Effects Using Low-Rank Estimations.” arXiv:1812.08089 [Math, Stat].
Chernozhukov, Newey, and Singh. 2018. Learning L2 Continuous Regression Functionals via Regularized Riesz Representers.” arXiv:1809.05224 [Econ, Math, Stat].
Chikuse. 2003. High Dimensional Asymptotic Theorems.” In Statistics on Special Manifolds.
Georgi. 2022. Physics in a Diverse World or A Spherical Cow Model of Physics Talent.” arXiv:2203.09485 [Hep-Ph, Physics:hep-Th, Physics:physics].
Gorban, Tyukin, and Romanenko. 2016. The Blessing of Dimensionality: Separation Theorems in the Thermodynamic Limit.” arXiv:1610.00494 [Cs, Stat].
Gribonval, Blanchard, Keriven, et al. 2017. Compressive Statistical Learning with Random Feature Moments.” arXiv:1706.07180 [Cs, Math, Stat].
Gui, and Li. 2005. Penalized Cox Regression Analysis in the High-Dimensional and Low-Sample Size Settings, with Applications to Microarray Gene Expression Data.” Bioinformatics.
Hall, and Li. 1993. On Almost Linearity of Low Dimensional Projections from High Dimensional Data.” The Annals of Statistics.
Javanmard, and Montanari. 2014. Confidence Intervals and Hypothesis Testing for High-Dimensional Regression.” Journal of Machine Learning Research.
Kaneko, Fiori, and Tanaka. 2013. Empirical Arithmetic Averaging Over the Compact Stiefel Manifold.” IEEE Transactions on Signal Processing.
Kar, and Karnick. 2012. Random Feature Maps for Dot Product Kernels.” In Artificial Intelligence and Statistics.
Müller, and van de Geer. 2015. Censored Linear Model in High Dimensions: Penalised Linear Regression on High-Dimensional Data with Left-Censored Response Variable.” TEST.
Oertel. 2020. Grothendieck’s Inequality and Completely Correlation Preserving Functions – a Summary of Recent Results and an Indication of Related Research Problems.”
Rigollet, and Hütter. 2019. High Dimensional Statistics.
Tang, and Reid. 2021. Laplace and Saddlepoint Approximations in High Dimensions.” arXiv:2107.10885 [Math, Stat].
Uematsu. 2015. Penalized Likelihood Estimation in High-Dimensional Time Series Models and Its Application.” arXiv:1504.06706 [Math, Stat].
van de Geer. 2014a. Worst Possible Sub-Directions in High-Dimensional Models.” In arXiv:1403.7023 [Math, Stat].
———. 2014b. Statistical Theory for High-Dimensional Models.” arXiv:1409.8557 [Math, Stat].
van de Geer, Bühlmann, Ritov, et al. 2014. On Asymptotically Optimal Confidence Regions and Tests for High-Dimensional Models.” The Annals of Statistics.
Veitch, and Roy. 2015. The Class of Random Graphs Arising from Exchangeable Random Measures.” arXiv:1512.03099 [Cs, Math, Stat].
Vershynin. 2015. Estimation in High Dimensions: A Geometric Perspective.” In Sampling Theory, a Renaissance: Compressive Sensing and Other Developments. Applied and Numerical Harmonic Analysis.
———. 2018. High-Dimensional Probability: An Introduction with Applications in Data Science.
Wainwright, Martin J. 2014. Structured Regularizers for High-Dimensional Problems: Statistical and Computational Issues.” Annual Review of Statistics and Its Application.
Wainwright, Martin. 2019. High-Dimensional Statistics: A Non-Asymptotic Viewpoint. Cambridge Series in Statistical and Probabilistic Mathematics 48.
Wright, and Ma. 2022. High-dimensional data analysis with low-dimensional models: Principles, computation, and applications.
Zhang, and Zhang. 2014. Confidence Intervals for Low Dimensional Parameters in High Dimensional Linear Models.” Journal of the Royal Statistical Society: Series B (Statistical Methodology).