Provably Feedback-Efficient Reinforcement Learning via Active Reward Learning.
Dingwen KongLin YangPublished in: NeurIPS (2022)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- learning problems
- supervised learning
- partially observable environments
- learning systems
- multi agent
- markov decision processes
- function approximation
- optimal control
- eligibility traces
- inverse reinforcement learning
- evolutionary learning
- reinforcement learning methods
- learning agents
- learning agent
- learning capabilities
- reinforcement learning algorithms
- action selection
- neural network
- knowledge acquisition
- state space
- mobile robot
- dynamic programming
- e learning