Open Problem: Kernel methods on manifolds and metric spaces. What is the probability of a positive definite geodesic exponential kernel?
Aasa FeragenSøren HaubergPublished in: COLT (2016)
Keyphrases
- positive definite
- metric space
- kernel methods
- kernel function
- euclidean space
- kernel matrix
- reproducing kernel hilbert space
- high dimensional
- geodesic distance
- probability measures
- feature space
- similarity search
- positive definite kernels
- support vector
- low dimensional
- kernel learning
- kernel machines
- riemannian manifolds
- graph kernels
- kernel pca
- distance function
- learning problems
- range queries
- input space
- hilbert space
- support vector machine
- point sets
- vector space
- machine learning
- support vectors
- finite dimensional
- covariance matrix
- high dimensional data
- data sets
- knn
- data mining