Policy Sharing Using Aggregation Trees for ${Q}$ -Learning in a Continuous State and Action Spaces.
Yu-Jen ChenWei-Cheng JiangMing-Yi JuKao-Shing HwangPublished in: IEEE Trans. Cogn. Dev. Syst. (2020)
Keyphrases
- continuous state and action spaces
- action selection
- reinforcement learning
- continuous state
- policy search
- function approximators
- markov decision problems
- optimal policy
- action space
- control policies
- state action
- partially observable markov decision processes
- markov decision process
- decision making
- function approximation
- temporal difference
- state dependent
- reinforcement learning algorithms
- neural network
- decision processes
- rl algorithms
- linear programming
- reward function
- robot navigation
- model free
- solving complex
- heuristic search
- policy gradient
- machine learning