Tighter Generalization Bounds for Matrix Completion Via Factorization Into Constrained Matrices.
Ken-ichiro MoridomiKohei HatanoEiji TakimotoPublished in: IEICE Trans. Inf. Syst. (2018)
Keyphrases
- matrix completion
- generalization bounds
- low rank
- data dependent
- learning theory
- generalization ability
- matrix factorization
- upper bound
- model selection
- vc dimension
- singular values
- ranking algorithm
- singular value decomposition
- linear classifiers
- learning problems
- low rank matrix
- convex optimization
- collaborative filtering
- missing data
- linear combination
- semi supervised
- lower bound
- high order
- high dimensional data
- incomplete data
- machine learning
- reinforcement learning
- expectation maximization
- kernel matrix
- objective function