MEPE: A Minimalist Ensemble Policy Evaluation Operator for Deep Reinforcement Learning.
Qiang HeXinwen HouPublished in: ICASSP (2024)
Keyphrases
- policy evaluation
- reinforcement learning
- temporal difference
- least squares
- model free
- markov decision processes
- function approximation
- monte carlo
- policy iteration
- learning algorithm
- td learning
- state space
- reinforcement learning algorithms
- random forests
- variance reduction
- multi agent
- semi parametric
- optimal policy
- ensemble methods
- feature selection
- evaluation function
- function approximators
- training data
- machine learning
- partially observable markov decision processes
- partially observable
- regression problems
- action selection
- dynamic programming
- supervised learning
- cross validation