Dealing With Changes: Resilient Routing via Graph Neural Networks and Multi-Agent Deep Reinforcement Learning.
Sai Shreyas BhavanasiLorenzo PapponeFlavio EspositoPublished in: IEEE Trans. Netw. Serv. Manag. (2023)
Keyphrases
- reinforcement learning
- multi agent
- neural network
- graph representation
- function approximation
- function approximators
- state space
- cooperative
- pattern recognition
- weighted graph
- markov decision processes
- reinforcement learning algorithms
- fuzzy logic
- multi agent systems
- routing problem
- graph model
- random walk
- back propagation
- single agent
- artificial neural networks
- multiagent systems
- graph theory
- network topology
- temporal difference
- directed graph
- connected components
- learning capabilities
- structured data
- learning process
- machine learning
- network topologies
- learning agents
- directed acyclic graph
- reinforcement learning agents
- recurrent neural networks
- graph structure
- bipartite graph
- neural nets
- routing algorithm
- ad hoc networks
- genetic algorithm
- learning algorithm
- traffic signal control
- optimal control
- multi agent reinforcement learning
- supervised learning
- sufficient conditions
- optimal policy
- self organizing maps
- autonomous agents
- neural network model
- graph theoretic
- model free
- spanning tree