Composable Geometric Motion Policies using Multi-Task Pullback Bundle Dynamical Systems.
Andrew BylardRiccardo BonalliMarco PavonePublished in: CoRR (2021)
Keyphrases
- dynamical systems
- multi task
- multi task learning
- predictive state representations
- partially observable markov decision processes
- learning tasks
- multi class
- human motion
- linear dynamical systems
- state space
- transfer learning
- feature selection
- learning problems
- nonlinear dynamical systems
- optimal policy
- gaussian processes
- information gain
- optical flow
- image processing
- intravascular ultrasound
- feature space
- dynamic programming
- high order
- image classification