Combining Deep Reinforcement Learning With Graph Neural Networks for Optimal VNF Placement.
Penghao SunJulong LanJunfei LiZehua GuoYuxiang HuPublished in: IEEE Commun. Lett. (2021)
Keyphrases
- neural network
- reinforcement learning
- dynamic programming
- optimal control
- pattern recognition
- fuzzy logic
- state space
- training process
- machine learning
- min sum
- solution path
- control policy
- neural nets
- learning algorithm
- optimal solution
- back propagation
- structured data
- infinite horizon
- approximate dynamic programming
- learning process
- reinforcement learning algorithms
- action selection
- graph theoretic
- model free
- undirected graph
- worst case
- function approximation
- optimal policy
- recurrent neural networks
- directed graph