Solving Robotic Manipulation With Sparse Reward Reinforcement Learning Via Graph-Based Diversity and Proximity.
Zhenshan BingHongkuan ZhouRui LiXiaojie SuFabrice O. MorinKai HuangAlois C. KnollPublished in: IEEE Trans. Ind. Electron. (2023)
Keyphrases
- reinforcement learning
- manipulation tasks
- state space
- function approximation
- partially observable environments
- markov decision problems
- markov decision processes
- eligibility traces
- temporal difference
- reinforcement learning algorithms
- solve complex tasks
- real robot
- learning algorithm
- model free
- machine learning
- reinforcement learning agents
- optimal policy
- compressive sensing
- graph model
- real time
- robot control
- control policy
- learning process
- reward function
- markov decision process
- average reward
- reinforcement learning methods
- robotic arm
- supervised learning
- high dimensional
- robotic control
- robotic systems