Learning active manipulation to target shapes with model-free, long-horizon deep reinforcement learning.
Matias SivertsvikKirill SumskiyEkrem MisimiPublished in: ICRA (2024)
Keyphrases
- reinforcement learning
- model free
- reinforcement learning methods
- function approximation
- reinforcement learning algorithms
- learning algorithm
- learning process
- rl algorithms
- temporal difference
- supervised learning
- temporal difference learning
- state space
- policy gradient
- learning tasks
- markov decision processes
- prior knowledge
- transfer learning