Designing Deep Reinforcement Learning for Human Parameter Exploration.
Hugo ScurtoBavo Van KerrebroeckBaptiste CaramiauxFrédéric BevilacquaPublished in: ACM Trans. Comput. Hum. Interact. (2021)
Keyphrases
- reinforcement learning
- active exploration
- guided exploration
- action selection
- machine learning
- learning process
- model based reinforcement learning
- exploration strategy
- model free
- human subjects
- learning algorithm
- autonomous learning
- temporal difference
- parameter space
- optimal policy
- e learning
- function approximation
- human behavior
- human interaction
- human robot interaction
- parameter settings
- input parameters
- parameter values
- human activities
- markov decision processes
- evolutionary algorithm