Reinforcement Learning of Impedance Policies for Peg-in-Hole Tasks: Role of Asymmetric Matrices.
Shir KozlovskyElad NewmanMiriam ZacksenhousePublished in: IEEE Robotics Autom. Lett. (2022)
Keyphrases
- reinforcement learning
- optimal policy
- reinforcement learning agents
- policy search
- tabula rasa
- transfer learning
- learning algorithm
- function approximation
- singular value decomposition
- markov decision processes
- robotic systems
- markov decision process
- multiagent reinforcement learning
- hierarchical reinforcement learning
- state space
- policy gradient methods
- machine learning
- multi task
- reward function
- reinforcement learning algorithms
- complex domains
- multi agent reinforcement learning