A Comparison of Spectral and Spatial Graph Convolutional Neural Network Kernels Using GraphSAGE-Sparse.
Michael EydenbergMark PlaggeSiva RajamanickamPublished in: IPDPS Workshops (2023)
Keyphrases
- convolutional neural network
- face detection
- graph kernels
- spatial frequency
- spatial information
- spatial and temporal
- spectral signatures
- spatio temporal
- random walk
- normalized cut
- structured data
- kernel methods
- spatial data
- graph model
- laplacian matrix
- spectral decomposition
- neural network
- multiple kernel
- feature space
- spectral analysis
- high dimensional
- weighted graph
- graph theory
- graph partitioning
- spectral methods
- color reproduction
- linear combination
- detection method
- kernel matrices
- kernel function
- directed graph
- sparse matrix
- bipartite graph
- hyperspectral
- kernel machines
- multiple kernel learning
- graph representation