Direct Policy Search Reinforcement Learning for Robot Control.
Andres El-FakdiMarc CarrerasNarcís PalomerasPublished in: CCIA (2005)
Keyphrases
- robot control
- direct policy search
- reinforcement learning
- mountain car
- function approximation
- state space
- learning algorithm
- control problems
- reinforcement learning algorithms
- model free
- subsumption architecture
- mobile robot
- dynamic programming
- markov decision processes
- machine learning
- temporal difference
- optimal policy
- multi agent
- autonomous robots
- action selection
- optimal control
- learning process
- function approximators
- reinforcement learning methods
- data mining
- transfer learning
- learning capabilities
- learning agent
- artificial neural networks