Low-Rank Univariate Sum of Squares Has No Spurious Local Minima.
Benoît LegatChenyang YuanPablo A. ParriloPublished in: CoRR (2022)
Keyphrases
- low rank
- linear combination
- missing data
- convex optimization
- matrix factorization
- singular value decomposition
- matrix completion
- semi supervised
- low rank matrix
- high order
- rank minimization
- high dimensional data
- kernel matrix
- robust principal component analysis
- matrix decomposition
- singular values
- trace norm
- collaborative filtering
- simulated annealing
- data matrix
- low rank approximation
- minimization problems
- small number
- low rank representation
- image classification
- higher order
- affinity matrix
- data mining
- data sets