SmartTRO: Optimizing topology robustness for Internet of Things via deep reinforcement learning with graph convolutional networks.
Yabin PengCaixia LiuShuxin LiuYuchen LiuYiteng WuPublished in: Comput. Networks (2022)
Keyphrases
- reinforcement learning
- small world
- fully connected
- deep learning
- edge weights
- complex networks
- social networks
- network structure
- directed graph
- average degree
- graph structure
- scale free
- weighted graph
- dynamic networks
- bipartite graph
- function approximation
- community discovery
- topological information
- state space
- protein interaction networks
- graph structures
- directed edges
- future internet
- reinforcement learning algorithms
- power law
- overlapping communities
- multi agent
- mobile devices
- random walk
- graph model
- graph mining
- community detection
- graph representation
- markov decision processes
- ubiquitous computing
- social graphs
- dynamic programming
- degree distribution
- interconnection networks
- structured data
- connected components
- community structure
- graph theory
- key technologies
- random graphs
- spanning tree
- undirected graph
- physical world
- graph theoretic
- model free