DeepMimic: example-guided deep reinforcement learning of physics-based character skills.
Xue Bin PengPieter AbbeelSergey LevineMichiel van de PannePublished in: ACM Trans. Graph. (2018)
Keyphrases
- reinforcement learning
- state space
- function approximation
- deformable models
- optimal policy
- learning process
- machine learning
- markov decision processes
- learning algorithm
- multi agent reinforcement learning
- handwritten characters
- learning environment
- deep learning
- reinforcement learning algorithms
- temporal difference
- dynamic programming
- chinese characters
- action selection
- real world
- information systems
- multi agent
- conceptual understanding
- supervised learning
- temporal difference learning
- reinforcement learning methods
- communication skills
- associative learning