Improving Reinforcement Learning with Confidence-Based Demonstrations.
Zhaodong WangMatthew E. TaylorPublished in: IJCAI (2017)
Keyphrases
- reinforcement learning
- function approximation
- confidence level
- model free
- machine learning
- reinforcement learning algorithms
- optimal policy
- real world
- learning algorithm
- multi agent reinforcement learning
- confidence measure
- high confidence
- robotic control
- temporal difference
- state space
- robot control
- learning classifier systems
- mobile robot
- database
- dynamic programming
- learning process
- multi agent systems
- multi agent
- multiscale
- bayesian networks
- image sequences
- information systems
- real time