Verifiably Safe Reinforcement Learning with Probabilistic Guarantees via Temporal Logic.
Hanna KrasowskiPrithvi AkellaAaron D. AmesMatthias AlthoffPublished in: CoRR (2022)
Keyphrases
- temporal logic
- reinforcement learning
- model checking
- modal logic
- satisfiability problem
- verification method
- concurrent systems
- computation tree logic
- mazurkiewicz traces
- temporal knowledge
- belief revision
- state space
- automata theoretic
- model checker
- formal specification language
- dynamic programming
- transition systems
- linear temporal logic
- probabilistic databases
- reactive systems
- markov decision processes
- domain specific
- search algorithm