On RKHS Choices for Assessing Graph Generators via Kernel Stein Statistics.
Moritz WeckbeckerWenkai XuGesine ReinertPublished in: CoRR (2022)
Keyphrases
- reproducing kernel hilbert space
- kernel methods
- kernel function
- graph kernels
- positive definite
- loss function
- reproducing kernel
- graph representation
- structured data
- kernel matrix
- special case
- random projections
- graph structure
- graph model
- weighted graph
- machine learning
- similarity function
- learning theory
- input space
- random walk
- dimensionality reduction
- hilbert space
- gaussian process
- graph mining
- graph theory
- gaussian processes
- euclidean space
- regularized least squares
- feature space