Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation.
Juncheng LiXin WangSiliang TangHaizhou ShiFei WuYueting ZhuangWilliam Yang WangPublished in: CVPR (2020)
Keyphrases
- reinforcement learning
- supervised learning
- unsupervised learning
- function approximation
- reinforcement learning algorithms
- meta level
- semi supervised
- model free
- data driven
- robot navigation
- state space
- navigation systems
- dynamic programming
- multi agent
- autonomous vehicles
- multi agent reinforcement learning
- temporal difference learning
- lifelong learning
- policy search
- robotic control
- optimal policy
- learning algorithm
- meta reasoning
- situated agents
- embodied cognition
- autonomous navigation
- temporal difference
- action selection
- information space
- active learning
- training data