Compression, inversion, and approximate PCA of dense kernel matrices at near-linear computational complexity.
Florian SchäferTimothy John SullivanHouman OwhadiPublished in: CoRR (2017)
Keyphrases
- linear computational complexity
- kernel matrices
- principal component analysis
- kernel methods
- kernel matrix
- low rank
- kernel pca
- linear complexity
- kernel function
- principal components analysis
- principal components
- feature extraction
- feature space
- linear combination
- image compression
- reproducing kernel hilbert space
- dimensionality reduction
- compression scheme
- kernel learning
- computational complexity
- compression ratio
- singular value decomposition
- high dimensional
- support vector machine
- generalization bounds
- face recognition
- convex optimization
- random projections
- neural network
- multiple kernel learning
- covariance matrix
- training samples
- support vector