Structured Cooperative Reinforcement Learning With Time-Varying Composite Action Space.
Wenhao LiXiangfeng WangBo JinDijun LuoHongyuan ZhaPublished in: IEEE Trans. Pattern Anal. Mach. Intell. (2022)
Keyphrases
- action space
- reinforcement learning
- cooperative
- state space
- state and action spaces
- markov decision processes
- real valued
- continuous state
- multi agent
- reinforcement learning methods
- single agent
- stochastic processes
- control policies
- action selection
- multi agent systems
- function approximators
- state action
- continuous state spaces
- function approximation
- partially observable
- reinforcement learning algorithms
- model free
- policy search
- optimal policy
- dynamic programming
- machine learning
- learning algorithm
- policy iteration
- control problems
- markov decision process
- markov decision problems
- temporal difference
- state variables
- latent variables
- transfer learning