DQN-TAMER: Human-in-the-Loop Reinforcement Learning with Intractable Feedback.
Riku ArakawaSosuke KobayashiYuya UnnoYuta TsuboiShin-ichi MaedaPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- feedback loop
- real time
- np complete
- multi agent
- human interaction
- function approximation
- learning algorithm
- human experts
- human operators
- reinforcement learning algorithms
- human users
- human subjects
- learning problems
- temporal difference learning
- motor skills
- tutorial dialogue
- user engagement
- creative problem solving
- temporal difference
- optimal control
- optimal policy
- human computer interaction
- computational complexity
- optimal solution
- data sets