SENTINEL: Taming Uncertainty with Ensemble-based Distributional Reinforcement Learning.
Hannes ErikssonDebabrota BasuMina AlibeigiChristos DimitrakakisPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- ensemble methods
- learning algorithm
- reinforcement learning algorithms
- state space
- function approximation
- ensemble learning
- partial observability
- neural network
- training set
- co occurrence
- optimal policy
- markov decision processes
- incomplete information
- possibility theory
- expected utility
- uncertain data
- temporal difference
- temporal difference learning
- random forest
- base classifiers
- data sets
- active learning
- random forests
- model free
- conditional probabilities
- artificial neural networks
- multi agent
- function approximators
- feature selection
- machine learning
- robotic control