Sampling-based Exploration for Reinforcement Learning of Dexterous Manipulation.
Gagan KhandateSiqi ShangEric T. ChangTristan Luca SaidiJohnson AdamsMatei T. CiocarliePublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- manipulation tasks
- active exploration
- motion planning
- exploration strategy
- action selection
- exploration exploitation
- haptic feedback
- model based reinforcement learning
- autonomous learning
- object manipulation
- markov decision processes
- function approximation
- humanoid robot
- state space
- monte carlo
- learning algorithm
- temporal difference
- model free
- temporal difference learning
- real robot
- exploration exploitation tradeoff
- adaptive sampling
- balancing exploration and exploitation
- reinforcement learning algorithms
- robot navigation
- optimal policy
- finite element analysis
- unknown environments
- multi agent
- partially observable
- learning process
- active learning
- vision system
- machine learning
- genetic algorithm
- robotic control
- dynamic programming
- supervised learning
- human activities
- reinforcement learning methods
- robot control