Metrics and continuity in reinforcement learning.
Charline Le LanMarc G. BellemarePablo Samuel CastroPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- function approximation
- temporal difference
- state space
- machine learning
- similarity metrics
- reinforcement learning algorithms
- optimal policy
- learning algorithm
- evaluation metrics
- model free
- temporal difference learning
- learning process
- multi agent
- multi agent reinforcement learning
- robotic control
- sufficient conditions
- learning environment
- markov decision processes
- evaluation criteria
- decision trees
- real world
- reinforcement learning methods
- direct policy search