Learning Synthetic Environments and Reward Networks for Reinforcement Learning.
Fabio FerreiraThomas NierhoffAndreas SaelingerFrank HutterPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- eligibility traces
- learning process
- learning algorithm
- supervised learning
- learning systems
- temporal difference learning
- neural network
- active learning
- reinforcement learning algorithms
- function approximation
- real world
- learning agents
- policy gradient
- learning problems
- autonomous learning
- evolutionary learning
- connectionist networks
- actor critic
- partially observable environments
- reinforcement learning methods
- learning agent
- temporal difference
- model free
- autonomous robots
- transfer learning
- state space
- machine learning