Rank-One Measurements of Low-Rank PSD Matrices Have Small Feasible Sets
arXiv:2012.09768
Abstract
We study the role of the constraint set in determining the solution to low-rank, positive semidefinite (PSD) matrix sensing problems. The setting we consider involves rank-one sensing matrices: In particular, given a set of rank-one projections of an approximately low-rank PSD matrix, we characterize the radius of the set of PSD matrices that satisfy the measurements. This result yields a sampling rate to guarantee singleton solution sets when the true matrix is exactly low-rank, such that the choice of the objective function or the algorithm to be used is inconsequential in its recovery. We discuss applications of this contribution and compare it to recent literature regarding implicit regularization for similar problems. We demonstrate practical implications of this result by applying conic projection methods for PSD matrix recovery without incorporating low-rank regularization.
arXiv comment22 pages, 3 figures
References in corpus (13)
- Distributed Representations of Words and Phrases and their Compositionality
- Guaranteed Minimum-Rank Solutions of Linear Matrix Equations via Nuclear Norm Minimization
- Quantum state tomography via compressed sensing
- An overview of low-rank matrix recovery from incomplete observations
- Quantum Tomography via Compressed Sensing: Error Bounds, Sample Complexity, and Efficient Estimators
- Self-Calibration and Biconvex Compressive Sensing
- A Unique "Nonnegative" Solution to an Underdetermined System: from Vectors to Matrices
- Blind Identification of Graph Filters
- Implicit Regularization in Deep Learning May Not Be Explainable by Norms
- Low-Rank Positive Semidefinite Matrix Recovery from Corrupted Rank-One Measurements
- Smoothed analysis for low-rank solutions to semidefinite programs in quadratic penalty form
- Provable Burer-Monteiro factorization for a class of norm-constrained matrix problems
- The Mixing method: low-rank coordinate descent for semidefinite programming with diagonal constraints