Fidelity-Induced Interpretable Policy Extraction for Reinforcement Learning.
Xiao LiuWubing ChenMao TanPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- action selection
- partially observable environments
- action space
- state space
- function approximators
- function approximation
- reward function
- policy iteration
- partially observable
- approximate dynamic programming
- markov decision processes
- reinforcement learning problems
- markov decision problems
- decision problems
- reinforcement learning algorithms
- policy evaluation
- actor critic
- state action
- control policy
- state dependent
- neural network
- partially observable domains
- learning algorithm
- multi agent
- model free
- dynamic programming
- information extraction
- policy gradient
- continuous state
- optimal control
- control policies
- learning problems
- classification rules
- least squares
- continuous state spaces
- knowledge extraction
- finite state
- state and action spaces
- infinite horizon
- partially observable markov decision processes