Datenbasierte Optimalsteuerung mit neuronalen Netzen und dateneffizientem Reinforcement Learning.
Thomas A. RunklerSteffen UdluftSiegmund DüllPublished in: Autom. (2012)
Keyphrases
- reinforcement learning
- function approximation
- state space
- learning algorithm
- learning process
- optimal policy
- test set
- markov decision processes
- model free
- temporal difference
- reinforcement learning algorithms
- markov chain
- massachusetts institute of technology
- transition model
- multi agent reinforcement learning
- evolutionary learning
- stochastic approximation
- learning agent
- control problems
- optimal control
- machine learning
- supervised learning
- least squares
- image sequences