Reinforcement learning using continuous states and interactive feedback.
Angel AyalaClaudio HenríquezFrancisco CruzPublished in: APPIS (2019)
Keyphrases
- reinforcement learning
- action space
- perceptual aliasing
- state space
- computer graphics
- state action
- state abstraction
- user friendly
- function approximation
- continuous domains
- multi agent
- user interaction
- robotic control
- continuous state
- initial state
- continuous state and action spaces
- state transition
- user feedback
- relevance feedback
- machine learning
- markov decision problems
- actor critic
- continuous state spaces
- markov decision processes
- virtual reality
- optimal policy
- fitted q iteration