Offline Reinforcement Learning with Imputed Rewards.
Carlo RomeoAndrew D. BagdanovPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- markov decision processes
- function approximation
- missing data
- state space
- reinforcement learning algorithms
- real time
- model free
- multi agent
- machine learning
- reward function
- missing values
- learning process
- hidden state
- dynamic programming
- learning algorithm
- reinforcement learning methods
- optimal policy
- supervised learning
- temporal difference learning
- bayesian networks
- temporal difference
- action space
- policy search
- optimal control
- learning problems
- transfer learning
- markov chain
- least squares
- genetic algorithm