Suboptimality bounds for trace-bounded SDPs enable a faster and scalable low-rank SDP solver SDPLR+.
Yufan HuangDavid F. GleichPublished in: CoRR (2024)
Keyphrases
- low rank
- kernel matrix
- convex optimization
- semidefinite programming
- missing data
- low rank matrix
- linear combination
- rank minimization
- matrix completion
- matrix factorization
- singular value decomposition
- semi supervised
- high order
- high dimensional data
- matrix decomposition
- eigendecomposition
- lower bound
- convex programming
- trace norm
- kernel learning
- singular values
- linear programming
- minimization problems
- robust principal component analysis
- missing values
- interior point methods
- sparse representation
- small number
- pairwise
- optimal solution