Improving reinforcement learning algorithms: towards optimal learning rate policies.
Othmane MounjidCharles-Albert LehallePublished in: CoRR (2019)
Keyphrases
- learning rate
- reinforcement learning algorithms
- total reward
- learning algorithm
- reinforcement learning
- policy search
- reward function
- state space
- markov decision processes
- convergence rate
- model free
- optimal policy
- dynamic programming
- multiagent reinforcement learning
- reinforcement learning problems
- weight vector
- temporal difference
- action selection
- markov decision process
- neural network
- average reward
- genetic algorithm
- optimal control
- cost function
- optimal solution