Advanced Self-Improving Ramp Metering Algorithm based on Multi-Agent Deep Reinforcement Learning.
Fuwen DengJiandong JinYu ShenYuchuan DuPublished in: ITSC (2019)
Keyphrases
- reinforcement learning
- multi agent
- learning algorithm
- dynamic programming
- preprocessing
- detection algorithm
- worst case
- computational cost
- optimization algorithm
- experimental evaluation
- np hard
- theoretical analysis
- high accuracy
- computational complexity
- optimal solution
- times faster
- segmentation algorithm
- improved algorithm
- expectation maximization
- significant improvement
- k means
- search space
- graph cuts
- path planning
- machine learning
- neural network