Computational and Statistical Boundaries for Submatrix Localization in a Large Noisy Matrix

Abstract

The interplay between computational efficiency and statistical accuracy in high-dimensional inference has drawn increasing attention in the literature. In this paper, we study computational and statistical boundaries for submatrix localization. Given one observation of (one or multiple non-overlapping) signal submatrix (of magnitude $\lambda$ and size $k_m\times k_n$) contaminated with a noise matrix (of size $m\times n$), we establish two transition thresholds for the signal to noise $\lambda /\sigma$ ratio in terms of $m$, $n$, $k_m$, and $k_n$. The first threshold, $\mathsf{SN}{\mathsf{R}}_{\mathsf{c}}$, corresponds to the computational boundary. Below this threshold, it is shown that no polynomial time algorithm can succeed in identifying the submatrix, under the \textit{hidden clique hypothesis}. We introduce adaptive linear time spectral algorithms that identify the submatrix with high probability when the signal strength is above the threshold $\mathsf{SN}{\mathsf{R}}_{\mathsf{c}}$. The second threshold, $\mathsf{SN}{\mathsf{R}}_{\mathsf{s}}$, captures the statistical boundary, below which no method can succeed with probability going to one in the minimax sense. The exhaustive search method successfully finds the submatrix above this threshold. The results show an interesting phenomenon that $\mathsf{SN}{\mathsf{R}}_{\mathsf{c}}$ is always significantly larger than $\mathsf{SN}{\mathsf{R}}_{\mathsf{s}}$, which implies an essential gap between statistical optimality and computational efficiency for submatrix localization.

37 pages, 1 figure

References in corpus (6)

• Finding large average submatrices in high dimensional data
• Statistical-Computational Tradeoffs in Planted Problems and Submatrix Localization with a Growing Number of Clusters and Submatrices
• Phase Transitions in Semidefinite Relaxations
• Lower bounds on the performance of polynomial-time algorithms for sparse linear regression
• Computational Lower Bounds for Sparse PCA
• Statistical and computational trade-offs in estimation of sparse principal components

Cited by in corpus (3)

• Computational Barriers to Estimation from Low-Degree Polynomials
• The Overlap Gap Property in Principal Submatrix Recovery
• Parallel Tempering for the planted clique problem