Bridging the Gap between Spatial and Spectral Domains: A Unified Framework for Graph Neural Networks.
Zhiqian ChenFanglan ChenLei ZhangTaoran JiKaiqun FuLiang ZhaoFeng ChenLingfei WuCharu AggarwalChang-Tien LuPublished in: ACM Comput. Surv. (2024)
Keyphrases
- neural network
- spatial frequency
- spectral decomposition
- artificial neural networks
- spatial information
- pattern recognition
- graph representation
- normalized cut
- random walk
- directed graph
- real world
- graph matching
- spatial data
- back propagation
- spatial distribution
- spectral characteristics
- spatio temporal
- graph theory
- connected components
- fuzzy logic
- graph structure
- spatial and temporal
- graph model
- multilayer perceptron
- spatial relationships
- spatial relations
- spectral methods
- training process
- laplacian matrix
- neural network model
- graph theoretic
- spectral signatures
- image segmentation
- graph laplacian
- graph partitioning
- undirected graph
- cross domain
- directed acyclic graph
- multi layer
- bipartite graph
- space time
- feed forward
- structured data
- fault diagnosis
- application domains