Reward-driven learning of sensorimotor laws and visual features

Link:
Autor/in:
Beteiligte Person:
  • Institute of Electrical and Electronics Engineers
Verlag/Körperschaft:
IEEE
Erscheinungsjahr:
2011
Medientyp:
Text
Schlagworte:
  • Reinforcement learning
  • Agents
  • Wages
  • Reinforcement Learning
  • Robots
  • Artificial Intelligence
  • Reinforcement learning
  • Agents
  • Wages
  • Reinforcement Learning
  • Robots
  • Artificial Intelligence
Beschreibung:
  • A frequently reoccurring task of humanoid robots is the autonomous navigation towards a goal position. Here we present a simulation of a purely vision-based docking behavior in a 3-D physical world. The robot learns sensorimotor laws and visual features simultaneously and exploits both for navigation towards its virtual target region. The control laws are trained using a two-layer network consisting of a feature (sensory) layer that feeds into an action (Q-value) layer. A reinforcement feedback signal (delta) modulates not only the action but at the same time the feature weights. Under this influence, the network learns interpretable visual features and assigns goal-directed actions successfully. This is a step towards investigating how reinforcement learning can be linked to visual perception.
Lizenz:
  • info:eu-repo/semantics/restrictedAccess
Quellsystem:
Forschungsinformationssystem der UHH

Interne Metadaten
Quelldatensatz
oai:www.edit.fis.uni-hamburg.de:publications/133ddfe9-bcb2-42ee-a7bc-c8ebd24401a6