Eligibility Propagation to Speed up Time Hopping for Reinforcement Learning.
Petar KormushevKohei NomotoFangyan DongKaoru HirotaPublished in: J. Adv. Comput. Intell. Intell. Informatics (2009)
Keyphrases
- reinforcement learning
- function approximation
- state space
- multi agent
- reinforcement learning algorithms
- learning algorithm
- model free
- robotic control
- database
- learning process
- real time
- multi agent reinforcement learning
- wave propagation
- temporal difference learning
- learning capabilities
- optimal policy
- real robot
- reinforcement learning methods
- action space
- learning agent
- robot control
- objective function
- temporal difference
- learning classifier systems
- dynamic programming
- transfer learning