Bayesian-Game-Based Fuzzy Reinforcement Learning Control for Decentralized POMDPs.
Rajneesh SharmaMatthijs T. J. SpaanPublished in: IEEE Trans. Comput. Intell. AI Games (2012)
Keyphrases
- reinforcement learning
- control problems
- multi agent
- markov decision processes
- fuzzy logic
- fuzzy sets
- optimal control
- control system
- control policy
- learning process
- state space
- function approximation
- distributed constraint optimization
- action selection
- optimal policy
- partially observable markov decision processes
- robot control
- adaptive control
- control strategies
- control method
- dynamic programming
- decentralized control
- reinforcement learning algorithms
- control policies
- cooperative
- dec pomdps
- model free
- fuzzy numbers
- reward function
- policy search
- fuzzy logic controller
- membership functions
- fuzzy controller
- continuous state
- policy gradient
- machine learning
- reinforcement learning methods
- policy iteration
- partially observable
- temporal difference
- peer to peer
- decision theory