Training Agents to Satisfy Timed and Untimed Signal Temporal Logic Specifications with Reinforcement Learning.
Nathaniel HamiltonPreston RobinetteTaylor T. JohnsonPublished in: SEFM (2022)
Keyphrases
- temporal logic
- reactive systems
- petri net
- reinforcement learning
- concurrent systems
- model checking
- multi agent
- transition systems
- model checker
- bounded model checking
- formal specification language
- action selection
- multi agent systems
- satisfiability problem
- learning agents
- modal logic
- verification method
- single agent
- formal specification
- cooperative
- computation tree logic
- epistemic logic
- mazurkiewicz traces
- partial observability
- temporal knowledge
- multiple agents
- agent receives
- state space
- learning agent
- dynamic constraints
- linear temporal logic
- logical framework
- belief revision
- heuristic search
- distributed systems
- learning algorithm