An Efficient and Globally Convergent Algorithm for $\ell_{p,q}$-$\ell_{r}$ Model in Group Sparse Optimization
arXiv:1904.018876 citations
AI Analysis
Provides a theoretically grounded algorithm for a broad class of group sparse optimization problems, benefiting researchers in signal processing and machine learning.
The authors propose InISSAPL, a proximally linearized algorithm for non-Lipschitz group sparse $\ell_{p,q}$-$\ell_r$ optimization, and prove its global convergence. Numerical experiments demonstrate its efficiency on synthetic and real-world data.
Group sparsity combines the underlying sparsity and group structure of the data in problems. We develop a proximally linearized algorithm InISSAPL for the non-Lipschitz group sparse $\ell_{p,q}$-$\ell_r$ optimization problem.