Autonomous reinforcement learning on raw visual input data in a real world application.
Sascha LangeMartin A. RiedmillerArne VoigtländerPublished in: IJCNN (2012)
Keyphrases
- input data
- reinforcement learning
- autonomous learning
- data sets
- multi agent
- visual perception
- function approximation
- state space
- visual information
- high level
- learning capabilities
- reinforcement learning algorithms
- visual features
- autonomous mental development
- raw data
- optimal policy
- feature space
- cooperative
- transfer learning
- database
- input image
- image data
- action selection
- temporal difference
- image sequences
- autonomous systems
- machine learning
- real time