Autonomous Learning of State Representations for Control: An Emerging Field Aims to Autonomously Learn State Representations for Reinforcement Learning Agents from Their Real-World Sensor Observations.
Wendelin BöhmerJost Tobias SpringenbergJoschka BoedeckerMartin A. RiedmillerKlaus ObermayerPublished in: Künstliche Intell. (2015)
Keyphrases
- reinforcement learning agents
- autonomous learning
- reinforcement learning
- dynamic environments
- control problems
- state abstraction
- data mining
- control system
- goal directed
- machine learning
- knowledge acquisition
- data sets
- learning algorithm
- multi agent environments
- evidential reasoning
- transfer learning
- knowledge base
- control strategy
- markov decision processes
- state space
- cooperative
- multi agent