Learning agent's spatial configuration from sensorimotor invariants.
Alban LaflaquièreJ. Kevin O'ReganSylvain ArgentieriBruno GasAlexander V. TerekhovPublished in: CoRR (2018)
Keyphrases
- learning agent
- spatial configuration
- reinforcement learning
- spatial distribution
- state space
- discriminative power
- solving problems
- learning algorithm
- keypoints
- spatial relationships
- learning capabilities
- spatial relations
- invariant features
- learning tasks
- reinforcement learning algorithms
- learning process
- object class
- spatial arrangement
- single agent
- neural network
- multi agent systems
- affine invariant
- dynamic environments
- spatial information
- reward function
- multi agent
- rigid transformation