IRNN for solving the following nonconvex nonsmooth low-rank minimization problem
where is continuous, concave and monotonically increasing on
, and
has Lipschitz continuous gradient.
References
- Nonconvex Nonsmooth Low-Rank Minimization via Iteratively Reweighted Nuclear Norm, Canyi Lu, Jinhui Tang, Shuicheng Yan, Zhouchen Lin, IEEE Transactions on Image Processing (TIP). 2016
- Generalized Nonconvex Nonsmooth Low-Rank Minimization, Canyi Lu, Jinhui Tang, Shuicheng Yan, Zhouchen Lin, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR) 2014: 4130-4137