Open the Black Box: Step-based Policy Updates for Temporally-Correlated Episodic Reinforcement Learning.
Ge LiHongyi ZhouDominik RothSerge ThilgesFabian OttoRudolf LioutikovGerhard NeumannPublished in: CoRR (2024)
Keyphrases
- black box
- reinforcement learning
- optimal policy
- policy search
- black boxes
- white box
- markov decision process
- integration testing
- action selection
- control policy
- partially observable environments
- state space
- test cases
- hybrid systems
- function approximation
- markov decision processes
- dynamic programming
- infinite horizon
- model free
- state action
- partially observable
- markov decision problems
- policy gradient
- state transition
- function approximators
- action space
- actor critic
- white box testing
- reward function
- reinforcement learning algorithms
- data sets
- temporal information
- artificial intelligence
- learning algorithm
- machine learning
- neural network