Discrete Uncertainty Quantification For Offline Reinforcement Learning.
José Luis PérezJavier CorrochanoJavier GarcíaRubén MajadasCristina Ibáñez-LlanoSergio PérezFernando FernándezPublished in: J. Artif. Intell. Soft Comput. Res. (2023)
Keyphrases
- reinforcement learning
- continuous state
- partial observability
- real time
- function approximation
- continuous state and action spaces
- inherent uncertainty
- continuous domains
- markov decision processes
- learning algorithm
- uncertain data
- temporal difference
- neural network
- control problems
- optimal control
- genetic algorithm
- finite number
- decision making
- reinforcement learning algorithms
- supervised learning
- state space
- robust optimization
- machine learning
- dynamic programming
- multi agent reinforcement learning
- continuous state spaces
- transition model
- learning process
- data sets