Virtual Network Embedding with Changeable Action Space: An Approach Based on Graph Neural Network and Reinforcement Learning.
Yawen TanJiadai WangJiajia LiuPublished in: ICC (2023)
Keyphrases
- action space
- reinforcement learning
- neural network
- state space
- state information
- markov decision processes
- function approximators
- state and action spaces
- continuous state
- real valued
- reinforcement learning methods
- continuous state spaces
- reinforcement learning algorithms
- action selection
- function approximation
- random walk
- multi agent
- state action
- finite state
- stochastic processes
- temporal difference
- link prediction
- network structure
- graphical models
- dynamic programming
- machine learning