loading page

A framework to generate sparsity-inducing regularizers for enhanced low-rank matrix completion
  • Zhi-Yong Wang ,
  • Hing Cheung so
Zhi-Yong Wang
City University of Hong Kong

Corresponding Author:[email protected]

Author Profile
Hing Cheung so
Author Profile


Applying half-quadratic optimization to loss functions can yield the corresponding regularizers, while these regularizers are usually not sparsity-inducing regularizers (SIRs). To solve this problem, we devise a framework to generate an SIR with closed-form proximity operator.  Besides, we specify our framework using several commonly-used loss functions, and produce the corresponding SIRs, which are then adopted as nonconvex rank surrogates for low-rank matrix completion. Furthermore, algorithms based on the alternating direction method of multipliers are developed. Extensive numerical results show the effectiveness of our methods in terms of recovery performance and runtime.