Perturbation-augmented Graph Convolutional Networks: A Graph Contrastive Learning architecture for effective node classification tasks.
Qihang GuoXibei YangFengjun ZhangTaihua XuPublished in: Eng. Appl. Artif. Intell. (2024)
Keyphrases
- graph structure
- fully connected
- reinforcement learning
- learning process
- learning algorithm
- directed graph
- path length
- graph representation
- random walk
- graph theory
- overlapping communities
- graph clustering
- finding the shortest path
- learning capabilities
- network architecture
- tree structure
- undirected graph
- graph partitioning
- directed acyclic graph
- bipartite graph
- small world
- shortest path
- graph structures
- unsupervised learning
- connectionist networks
- online learning
- active learning