Implications of Centralized and Distributed Multi-Agent Deep Reinforcement Learning in Dynamic Spectrum Access.
Abdikarim Mohamed IbrahimKok-Lim Alvin YauMee Hong LingPublished in: ISTT (2022)
Keyphrases
- multi agent
- reinforcement learning
- cooperative
- distributed environment
- peer to peer
- dynamic environments
- multi agent systems
- state space
- fully distributed
- single agent
- learning algorithm
- multi agent environments
- open distributed systems
- multiagent environments
- multiple autonomous
- geographically distributed
- multiple agents
- function approximation
- intelligent agents
- distributed systems
- model free
- markov decision processes
- heterogeneous agents
- team formation
- multi agent architecture
- access control
- uniform access
- transparent access
- neural network