Shaping Progressive Net of Reinforcement Learning for Policy Transfer with Human Evaluative Feedback.
Rongshun JuanJie HuangRandy GomezKeisuke NakamuraQixin ShaBo HeGuangliang LiPublished in: IROS (2021)
Keyphrases
- reinforcement learning
- optimal policy
- reward shaping
- transfer learning
- action selection
- markov decision process
- state and action spaces
- reinforcement learning problems
- markov decision processes
- policy search
- actor critic
- human operators
- reinforcement learning algorithms
- control policies
- human subjects
- function approximation
- state space
- markov decision problems
- reward signal
- machine learning
- partially observable markov decision processes
- model free
- human interaction
- dynamic programming
- multi agent
- optimal control
- human behavior