Researchers have developed a new optimization method called Controlled Block Proximal Linearized (CBPL) for complex, non-convex, and non-smooth problems. This method incorporates a control potential to ensure boundedness and convergence of variables and objective values. CBPL has been applied to various matrix factorization techniques, including nonnegative matrix factorization (NMF) and CP decomposition, demonstrating its effectiveness in achieving convergence and controlled scaling. AI
RANK_REASON The cluster contains a research paper detailing a new optimization method. [lever_c_demoted from research: ic=1 ai=0.4]
- Controlled Block Proximal Linearized (CBPL)
- CP decomposition
- CP models
- $\ell_0$ norm
- graph-regularized nonnegative matrix factorization (NMF)
- independent-core matrix tri-factorization (NMTF)
- KL property
- non-negative matrix factorization
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →