Rotational invariant estimator for general noisy matrices

Abstract

We investigate the problem of estimating a given real symmetric signal matrix $\text{C}$ from a noisy observation matrix $\text{M}$ in the limit of large dimension. We consider the case where the noisy measurement $\text{M}$ comes either from an arbitrary additive or multiplicative rotational invariant perturbation. We establish, using the Replica method, the asymptotic global law estimate for three general classes of noisy matrices, significantly extending previously obtained results. We give exact results concerning the asymptotic deviations (called overlaps) of the perturbed eigenvectors away from the true ones, and we explain how to use these overlaps to "clean" the noisy eigenvalues of $\text{M}$. We provide some numerical checks for the different estimators proposed in this paper and we also make the connection with some well known results of Bayesian statistics.

19 pages, 9 figures

References in corpus (1)

• On asymptotics of eigenvectors of large sample covariance matrix

Cited by in corpus (26)

• Cleaning large correlation matrices: tools from random matrix theory
• PCA meets RG
• Eigenvectors distribution and quantum unique ergodicity for deformed Wigner matrices
• Perturbative construction of mean-field equations in extensive-rank matrix factorization and denoising
• Statistical limits of dictionary learning: random matrix theory and the spectral replica method
• On the overlaps between eigenvectors of correlated random matrices
• Convergence of eigenvector empirical spectral distribution of sample covariance matrices
• Matrix factorization with neural networks
• Depth induces scale-averaging in overparameterized linear Bayesian neural networks
• Covariance matrix filtering with bootstrapped hierarchies
• Inferring hidden states in Langevin dynamics on large networks: Average case performance
• Singular Vectors of Sums of Rectangular Random Matrices and Optimal Estimators of High-Rank Signals: The Extensive Spike Model
• Matrix denoising: Bayes-optimal estimators via low-degree polynomials
• Optimal design of experiments by combining coarse and fine measurements
• Statistical diagonalization of a random biased Hamiltonian: the case of the eigenvectors
• Exploration of Superposition Theorem in Spectrum Space for Composite Event Analysis in an ADN
• On the variability of the sample covariance matrix under complex elliptical distributions
• Some observations on the ambivalent role of symmetries in Bayesian inference problems
• A Compound Decision Approach to Covariance Matrix Estimation
• Two short pieces around the Wigner problem
• The Wishart--Rosenzweig--Porter random matrix ensemble
• Functional Renormalization Group Approach for Signal Detection
• "Spectrally gapped" random walks on networks: a Mean First Passage Time formula
• Statistical physics of deep learning: Optimal learning of a multi-layer perceptron near interpolation
• Optimal generalisation and learning transition in extensive-width shallow neural networks near interpolation
• Statistical mechanics of extensive-width Bayesian neural networks near interpolation