Large deviations of extreme eigenvalues of generalized sample covariance matrices

Abstract

We present an analytical technique to compute the probability of rare events in which the largest eigenvalue of a random matrix is atypically large (i.e.\ the right tail of its large deviations). The results also transfer to the left tail of the large deviations of the smallest eigenvalue. The technique improves upon past methods by not requiring the explicit law of the eigenvalues, and we apply it to a large class of random matrices that were previously out of reach. In particular, we solve an open problem related to the performance of principal components analysis on highly correlated data, and open the way towards analyzing the high-dimensional landscapes of complex inference models. We probe our results using an importance sampling approach, effectively simulating events with probability as small as $10^{-100}$.

arXiv comment7 pages + 7 pages appendix. Updated version matching the published article

References in corpus (15)

• Level-Spacing Distributions and the Airy Kernel
• Statistical physics of inference: Thresholds and algorithms
• Random Matrix Theory and Chiral Symmetry in QCD
• Large Deviations of Extreme Eigenvalues of Random Matrices
• Replica Field Theory for Deterministic Models (II): A Non-Random Spin Glass with Glassy Behavior
• Large Deviations of the Maximum Eigenvalue for Wishart and Gaussian Random Matrices
• Mean-Field Equations for Spin Models with Orthogonal Interaction Matrices
• Large Deviations of the Maximum Eigenvalue in Wishart Random Matrices
• On the Large N Limit of the Itzykson-Zuber Integral
• On some integrals over the U(N) unitary group and their large N limit
• On large deviation properties of Erdos-Renyi random graphs
• High-precision simulation of the height distribution for the KPZ equation
• High-temperature Expansions and Message Passing Algorithms
• Multicanonical sampling of rare events in random matrices
• Random Matrix Model for Superconductors in a Magnetic Field

Cited by in corpus (1)

• Right large deviation principle for the top eigenvalue of the sum or product of invariant random matrices