Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning.
Eric VollenweiderMarko BjelonicVictor KlemmNikita RudinJoonho LeeMarco HutterPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- multi agent
- image sequences
- motion model
- motion patterns
- function approximation
- human motion
- motion field
- motion analysis
- motion estimation
- camera motion
- three dimensional
- optical flow
- prior knowledge
- humanoid robot
- model free
- dynamic programming
- feature points
- spatial and temporal
- learning process
- motion segmentation
- spatio temporal
- computational complexity
- motion capture
- motion tracking
- reinforcement learning algorithms