A Residual Bootstrap for High-Dimensional Regression with Near Low-Rank Designs.
Miles LopesPublished in: NIPS (2014)
Keyphrases
- low rank
- high dimensional
- high dimensional data
- low dimensional
- convex optimization
- singular value decomposition
- linear combination
- matrix factorization
- dimensionality reduction
- matrix completion
- missing data
- low rank matrix
- rank minimization
- regression model
- cross validation
- matrix decomposition
- data points
- semi supervised
- high dimensionality
- model selection
- high order
- similarity search
- kernel matrix
- variable selection
- data sets
- regularized regression
- nearest neighbor
- input space
- robust principal component analysis
- singular values
- non rigid structure from motion
- support vector
- sparse coding
- reproducing kernel hilbert space
- low rank matrices
- trace norm
- feature space
- pattern recognition
- machine learning
- manifold learning
- learning algorithm
- kernel matrices
- minimization problems
- data analysis
- contingency tables
- support vector machine
- missing values