Interpretable and Editable Programmatic Tree Policies for Reinforcement Learning.
Hector KohlerQuentin DelfosseRiad AkrourKristian KerstingPhilippe PreuxPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- control policies
- tree structure
- function approximation
- reward function
- policy gradient methods
- markov decision process
- reinforcement learning algorithms
- markov decision problems
- markov decision processes
- state space
- dynamic programming
- multi agent
- temporal difference
- control policy
- robotic control
- fitted q iteration
- tree construction
- continuous state
- reinforcement learning agents
- machine learning
- tree structures
- partially observable markov decision processes
- leaf nodes
- action space
- long run
- sufficient conditions
- tree models
- multi dimensional
- learning process
- association rules
- search algorithm
- learning algorithm