Marginalized Operators for Off-policy Reinforcement Learning.
Yunhao TangMark RowlandRémi MunosMichal ValkoPublished in: AISTATS (2022)
Keyphrases
- reinforcement learning
- function approximation
- dynamic programming
- temporal difference
- model free
- morphological operators
- reinforcement learning algorithms
- partially observable
- multi agent
- optimal control
- learning process
- temporal difference learning
- relational reinforcement learning
- action selection
- function approximators
- robotic control
- learning capabilities
- multi agent reinforcement learning
- real time
- evaluation function
- markov decision processes
- state space
- active learning
- decision trees
- machine learning
- data mining
- neural network