Scaled stochastic gradient descent for low-rank matrix completion.
Bamdev MishraRodolphe SepulchrePublished in: CDC (2016)
Keyphrases
- stochastic gradient descent
- matrix completion
- low rank
- matrix factorization
- least squares
- loss function
- missing data
- rank minimization
- collaborative filtering
- step size
- random forests
- negative matrix factorization
- kernel matrix
- recommender systems
- nonnegative matrix factorization
- online algorithms
- weight vector
- singular value decomposition
- regularization parameter
- multiple kernel learning
- missing values
- importance sampling
- singular values
- linear combination
- principal component analysis
- optical flow