A Rank-Corrected Procedure for Matrix Completion with Fixed Basis Coefficients
This addresses a bottleneck in matrix completion for applications like finance and quantum tomography, offering a significant but incremental improvement over existing methods.
The paper tackles low-rank matrix completion when certain basis coefficients are fixed, proposing a rank-corrected procedure that reduces recovery error by around 50% compared to nuclear norm methods, as shown in non-asymptotic bounds and numerical experiments.
For the problems of low-rank matrix completion, the efficiency of the widely-used nuclear norm technique may be challenged under many circumstances, especially when certain basis coefficients are fixed, for example, the low-rank correlation matrix completion in various fields such as the financial market and the low-rank density matrix completion from the quantum state tomography. To seek a solution of high recovery quality beyond the reach of the nuclear norm, in this paper, we propose a rank-corrected procedure using a nuclear semi-norm to generate a new estimator. For this new estimator, we establish a non-asymptotic recovery error bound. More importantly, we quantify the reduction of the recovery error bound for this rank-corrected procedure. Compared with the one obtained for the nuclear norm penalized least squares estimator, this reduction can be substantial (around 50%). We also provide necessary and sufficient conditions for rank consistency in the sense of Bach (2008). Very interestingly, these conditions are highly related to the concept of constraint nondegeneracy in matrix optimization. As a byproduct, our results provide a theoretical foundation for the majorized penalty method of Gao and Sun (2010) and Gao (2010) for structured low-rank matrix optimization problems. Extensive numerical experiments demonstrate that our proposed rank-corrected procedure can simultaneously achieve a high recovery accuracy and capture the low-rank structure.