Decoupling Meta-Reinforcement Learning with Gaussian Task Contexts and Skills.
Hongcai HeAnjie ZhuShuang LiangFeiyu ChenJie ShaoPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- input output
- function approximation
- learning algorithm
- multi agent
- model free
- reinforcement learning algorithms
- state space
- markov decision processes
- associative learning
- learning process
- robotic control
- maximum likelihood
- social context
- meta level
- machine learning
- function approximators
- multi agent reinforcement learning
- temporal difference
- optimal policy
- policy search
- neural network
- normal distribution
- gaussian distribution
- action selection
- gaussian mixture model
- web applications
- mobile robot
- technical skills
- knowledge domains
- gaussian function
- skill development
- dynamic programming