Unsupervised Reinforcement Learning for Transferable Manipulation Skill Discovery

Current reinforcement learning (RL) in robotics often experiences difficulty in generalizing to new downstream tasks due to the innate task-specific training paradigm. To alleviate it, unsupervised RL, a framework that pre-trains the agent in a task-agnostic manner without access to the task-specifi...

Full description

Bibliographic Details
Main Authors:	Cho, D. (Author), Kim, H.J (Author), Kim, J. (Author)
Format:	Article
Language:	English
Published:	Institute of Electrical and Electronics Engineers Inc. 2022
Subjects:	Deep learning Deep learning method Deep Learning Methods Job analysis Learning methods Machine learning for robot control Machine Learning for Robot Control Machine-learning Mutual information Mutual informations Personnel training Reinforcement learning Reinforcement Learning Reinforcement learnings Robotics Robots Robots control Space exploration Space explorations Space research Task analysis Training Transfer learning Transfer Learning Uncertainty Uncertainty analysis
Online Access:	View Fulltext in Publisher

Internet

View Fulltext in Publisher

Unsupervised Reinforcement Learning for Transferable Manipulation Skill Discovery

Internet

Similar Items