Deep Reinforcement Learning Method of Target Hunting for Multi-agents with Flocking Property.
Yujiao DongJin ChengQiqi WangPublished in: NCAA (1) (2023)
Keyphrases
- cost function
- segmentation method
- detection method
- multi agent
- high accuracy
- pairwise
- reinforcement learning
- prior knowledge
- significant improvement
- high precision
- synthetic data
- clustering method
- computationally efficient
- probabilistic model
- computational cost
- experimental evaluation
- dynamic programming
- data sets
- artificial neural networks
- preprocessing
- video sequences
- objective function
- similarity measure
- face recognition
- learning algorithm
- neural network