Benchmarking Offline Reinforcement Learning on Real-Robot Hardware.
Nico GürtlerSebastian BlaesPavel KolevFelix WidmaierManuel WuthrichStefan BauerBernhard SchölkopfGeorg MartiusPublished in: ICLR (2023)
Keyphrases
- temporal difference
- real robot
- reinforcement learning
- function approximation
- real time
- reinforcement learning algorithms
- hardware and software
- low cost
- simulated robot
- evolutionary robotics
- state space
- robot soccer
- multi agent
- learning algorithm
- hardware implementation
- mobile robot
- markov decision processes
- motion control
- manipulation tasks
- real environment
- machine learning
- robotic systems
- dynamical systems
- optimal policy
- real world
- data sets