Pre-training with non-expert human demonstration for deep reinforcement learning.
Gabriel Victor de la CruzYunshu DuMatthew E. TaylorPublished in: Knowl. Eng. Rev. (2019)
Keyphrases
- reinforcement learning
- human experts
- training process
- markov decision processes
- function approximation
- supervised learning
- deep architectures
- motor skills
- training set
- training algorithm
- training examples
- optimal policy
- training phase
- human behavior
- genetic algorithm
- state space
- hidden markov models
- active learning
- multi agent
- training data