Asymptotic Normality of Log Likelihood Ratio and Fundamental Limit of the Weak Detection for Spiked Wigner Matrices

TL;DR

This paper establishes the asymptotic Gaussian behavior of the log likelihood ratio in spiked Wigner models with non-Gaussian noise, identifying the fundamental detection limit and linking it to PCA performance.

Contribution

It proves the Gaussian convergence of the log likelihood ratio below a certain threshold and characterizes the fundamental detection limit for rank-one spiked models with non-Gaussian noise.

Findings

Log likelihood ratio converges to a Gaussian distribution below the detection threshold.

The detection threshold is optimal and aligns with PCA-based methods.

The results extend to asymmetric noise in IID models.

Abstract

We consider the problem of detecting the presence of a signal in a rank-one spiked Wigner model. For general non-Gaussian noise, assuming that the signal is drawn from the Rademacher prior, we prove that the log likelihood ratio (LR) of the spiked model against the null model converges to a Gaussian when the signal-to-noise ratio is below a certain threshold. The threshold is optimal in the sense that the reliable detection is possible by a transformed principal component analysis (PCA) above it. From the mean and the variance of the limiting Gaussian for the log-LR, we compute the limit of the sum of the Type-I error and the Type-II error of the likelihood ratio test. We also prove similar results for a rank-one spiked IID model where the noise is asymmetric but the signal is symmetric.