On Reward Distribution in Reinforcement Learning of Multi-Agent Surveillance Systems With Temporal Logic Specifications.
Keita TerashimaKoichi KobayashiYuh YamashitaPublished in: GCCE (2023)
Keyphrases
- temporal logic
- reinforcement learning
- surveillance system
- multi agent
- concurrent systems
- model checking
- transition systems
- bounded model checking
- reactive systems
- model checker
- formal specification language
- video surveillance
- state space
- modal logic
- formal specification
- reinforcement learning algorithms
- background subtraction
- learning algorithm
- mazurkiewicz traces
- reward function
- markov decision processes
- linear temporal logic
- dynamic programming
- traffic surveillance
- formal verification
- verification method
- optimal policy
- symbolic model checking
- machine learning
- computation tree logic
- communication networks
- belief revision
- sparse representation
- dynamical systems
- dynamic constraints
- learning agent
- markov decision process
- temporally extended
- ptz cameras
- control flow
- search algorithm
- single agent
- computer vision