Tight Dimensionality Reduction for Sketching Low Degree Polynomial Kernels.
Michela MeisterTamás SarlósDavid P. WoodruffPublished in: NeurIPS (2019)
Keyphrases
- low degree
- polynomial kernels
- dimensionality reduction
- high dimensional feature space
- kernel function
- feature space
- support vector machine
- gaussian kernels
- support vector
- lower bound
- high dimensional
- linearly separable
- upper bound
- high dimensional data
- principal component analysis
- feature selection
- input space
- high dimensionality
- low dimensional
- pattern recognition
- kernel pca
- kernel learning
- feature extraction
- uniform distribution
- worst case
- euclidean distance
- principal components
- data points
- data sets
- kernel methods
- objective function
- agnostic learning
- decision lists
- kernel matrix
- optimal solution
- multiple kernel learning
- metric learning
- svm classifier
- classification accuracy