Ensemble Bootstrapped Deep Deterministic Policy Gradient for Vision-Based Robotic Grasping.
Weiwei LiuLinpeng PengJunjie CaoXiaokuan FuYong LiuZaisheng PanJian YangPublished in: IEEE Access (2021)
Keyphrases
- policy gradient
- object manipulation
- vision system
- manipulation tasks
- real time
- robotic systems
- function approximation
- reinforcement learning
- actor critic
- gradient method
- optimal control
- model free reinforcement learning
- training data
- reinforcement learning algorithms
- mobile robot
- approximation methods
- training set
- learning algorithm
- variance reduction
- neural network
- markov chain
- single agent
- real robot
- feature selection