Intelligent Trainer for Dyna-Style Model-Based Deep Reinforcement Learning.
Linsen DongYuanlong LiXin ZhouYonggang WenKyle GuanPublished in: IEEE Trans. Neural Networks Learn. Syst. (2021)
Keyphrases
- reinforcement learning
- function approximation
- model free
- temporal difference learning
- temporal difference
- rl algorithms
- reinforcement learning algorithms
- reinforcement learning methods
- function approximators
- state space
- decision support
- intelligent systems
- multi agent reinforcement learning
- control problems
- learning capabilities
- dynamic programming
- machine learning
- markov decision processes
- optimal policy
- context aware
- learning algorithm
- game playing
- learning process
- policy gradient
- neural network