Judgmentally adjusted Q-values based on Q-ensemble for offline reinforcement learning.
Wenzhuo LiuShuying XiangTao ZhangYanan HanXingxing GuoYahui ZhangYue HaoPublished in: Neural Comput. Appl. (2024)
Keyphrases
- reinforcement learning
- learning algorithm
- function approximation
- real time
- neural network
- reinforcement learning algorithms
- ensemble learning
- artificial neural networks
- learning process
- state space
- temporal difference
- optimal policy
- random forests
- ensemble methods
- ensemble pruning
- parameter values
- markov decision processes
- supervised learning
- multi agent
- data sets