Edge-Varying Fourier Graph Networks for Multivariate Time Series Forecasting.
Kun YiQi ZhangLiang HuHui HeNing AnLongbing CaoZhendong NiuPublished in: CoRR (2022)
Keyphrases
- edge weights
- weighted graph
- undirected graph
- dynamic networks
- densely connected
- bipartite graph
- network analysis
- disjoint paths
- graph structures
- graph theory
- social networks
- small world
- graph representation
- spanning tree
- average degree
- highly connected
- fourier spectrum
- shortest path
- social graphs
- vertex set
- network structure
- graph layout
- directed acyclic graph
- connected components
- directed graph
- network size
- image reconstruction
- fully connected
- degree distribution
- edge detection
- strongly connected
- graph data
- artificial neural networks
- complex networks
- neural network
- edge information
- graph structure
- path length
- directed edges
- citation networks
- random walk
- random graphs
- community structure
- frequency domain
- graphical representation
- graph theoretic
- graph databases
- radon transform
- graph mining
- fourier transform