Less Is More: Refining Datasets for Offline Reinforcement Learning with Reward Machines.
Haoyuan SunFeng WuPublished in: AAMAS (2023)
Keyphrases
- reinforcement learning
- function approximation
- eligibility traces
- multi agent
- reinforcement learning algorithms
- machine learning
- learning algorithm
- function approximators
- transfer learning
- learning problems
- neural network
- uci machine learning repository
- learning agent
- temporal difference
- model free
- optimal control
- markov decision processes
- benchmark datasets
- average reward
- reward shaping
- partially observable environments
- action selection
- reward function
- synthetic and real datasets
- supervised learning
- policy search
- real time