Learning a Joint Low-Rank and Gaussian Model in Matrix Completion with Spectral Regularization and Expectation Maximization Algorithm.
Gang WuRatnesh KumarPublished in: BigData Congress (2018)
Keyphrases
- matrix completion
- low rank
- expectation maximization
- trace norm
- learning algorithm
- rank minimization
- em algorithm
- unsupervised learning
- norm minimization
- stochastic gradient descent
- linear combination
- low rank matrices
- missing data
- singular values
- k means
- supervised learning
- matrix factorization
- active learning
- maximum likelihood
- probabilistic model
- singular value decomposition
- convex optimization
- feature vectors
- high dimensional
- low rank matrix
- feature space
- similarity measure
- computer vision