Consistency of Interpolation with Laplace Kernels is a High-Dimensional Phenomenon.
Alexander RakhlinXiyu ZhaiPublished in: CoRR (2018)
Keyphrases
- high dimensional
- kernel function
- feature space
- similarity search
- low dimensional
- high dimensionality
- dimensionality reduction
- sparse data
- additive models
- variable selection
- gene expression data
- global consistency
- image interpolation
- high dimensional spaces
- input space
- high dimensional data
- nearest neighbor
- gaussian processes
- multi dimensional
- high dimensional problems
- support vector
- neural network
- parameter space
- kernel methods
- multiple kernel learning
- linear combination
- consistency checking
- feature vectors
- linear interpolation
- gaussian kernels
- interpolation algorithm