Successive Over-Relaxation ${Q}$ -Learning.
Chandramouli KamanchiRaghuram Bharadwaj DiddigiShalabh BhatnagarPublished in: IEEE Control. Syst. Lett. (2020)
Keyphrases
- reinforcement learning
- function approximation
- cooperative
- multi agent
- stochastic approximation
- state space
- learning algorithm
- action selection
- reinforcement learning algorithms
- optimal policy
- multi agent reinforcement learning
- probabilistic relaxation
- lognormal distribution
- bucket brigade
- iterative algorithms
- temporal difference learning
- potential field
- convex relaxation
- policy iteration
- state action
- fixed point
- reward function
- linear programming relaxation
- learning rate
- markov decision processes
- artificial neural networks
- objective function