A reinforcement learning method to improve the sweeping efficiency for an agent.
Naoto OhsakaDaisuke KitakoshiMasato SuzukiPublished in: GrC (2011)
Keyphrases
- reinforcement learning
- high efficiency
- detection method
- significant improvement
- model free
- highly efficient
- pairwise
- cost function
- experimental evaluation
- classification accuracy
- high accuracy
- objective function
- multi agent
- decision making
- preprocessing
- intelligent agents
- learning algorithm
- feature extraction
- segmentation method
- dynamic programming
- genetic algorithm
- computational complexity
- multi agent systems