Statistical properties of large data sets with linear latent features

Abstract

Analytical understanding of how low-dimensional latent features reveal themselves in large-dimensional data is still lacking. We study this by defining a linear latent feature model with additive noise constructed from probabilistic matrices, and analytically and numerically computing the statistical distributions of pairwise correlations and eigenvalues of the correlation matrix. This allows us to resolve the latent feature structure across a wide range of data regimes set by the number of recorded variables, observations, latent features and the signal-to-noise ratio. We find a characteristic imprint of latent features in the distribution of correlations and eigenvalues and provide an analytic estimate for the boundary between signal and noise even in the absence of a clear spectral gap.

arXiv comment20 pages, 6 figures

References in corpus (12)

• Deep learning for universal linear embeddings of nonlinear dynamics
• Dimensionality and dynamics in the behavior of C. elegans
• Introduction to Random Matrices - Theory and Practice
• Collective behavior of place and non-place neurons in the hippocampal network
• Coarse--graining, fixed points, and scaling in a large population of neurons
• Zipf's law and criticality in multivariate data without fine-tuning
• Eigenvalues and Singular Values of Products of Rectangular Gaussian Random Matrices
• Revealing the state space of turbulence using machine learning
• Latent dynamical variables produce signatures of spatiotemporal criticality in large biological systems
• Spectrum of deformed random matrices and free probability
• Spectral density of products of Wishart dilute random matrices. Part I: the dense case
• The Perturbative Resolvent Method: spectral densities of random matrix ensembles via perturbation theory

Cited by in corpus (5)

• Bayesian Renormalization
• Singular Vectors of Sums of Rectangular Random Matrices and Optimal Estimators of High-Rank Signals: The Extensive Spike Model
• Generative random latent features models and statistics of natural images
• Renormalization group for spectral collapse in random matrices with power-law variance profiles
• Distribution of singular values in large sample cross-covariance matrices