Dexterous Robotic Manipulation using Deep Reinforcement Learning and Knowledge Transfer for Complex Sparse Reward-based Tasks.
Qiang WangFrancisco Roldan SanchezRobert McCarthyDavid Cordova BulensKevin McGuinnessNoel E. O'ConnorManuel WüthrichFelix WidmaierStefan BauerStephen J. RedmondPublished in: CoRR (2022)
Keyphrases
- knowledge transfer
- reinforcement learning
- transfer learning
- manipulation tasks
- real robot
- transferring knowledge
- object manipulation
- learning tasks
- robot navigation
- previously learned
- state space
- multi task
- robotic systems
- transfer knowledge
- cross domain
- machine learning
- motion planning
- artificial intelligence
- reinforcement learning algorithms
- knowledge sharing
- real world
- learning algorithm
- active learning
- robot control
- mobile robot
- supervised learning
- collaborative filtering
- labeled data
- service robots
- text categorization
- optimal policy
- human robot interaction
- high dimensional
- markov decision processes
- decision making
- robotic arm
- autonomous robots
- learning process
- humanoid robot