Targeted Adversarial Attacks on Deep Reinforcement Learning Policies via Model Checking.
Dennis GrossThiago D. SimãoNils JansenGuillermo A. PérezPublished in: CoRR (2022)
Keyphrases
- model checking
- reinforcement learning
- optimal policy
- temporal logic
- markov decision processes
- markov decision process
- multi agent
- reachability analysis
- finite state
- model checker
- formal specification
- state space
- formal verification
- partially observable markov decision processes
- reward function
- symbolic model checking
- finite state machines
- temporal properties
- automated verification
- partial order reduction
- bounded model checking
- markov decision problems
- reinforcement learning algorithms
- epistemic logic
- transition systems
- computation tree logic
- asynchronous circuits
- dynamic programming
- timed automata
- concurrent systems
- decision problems
- learning algorithm
- pspace complete
- temporally extended
- verification method
- infinite horizon
- satisfiability problem
- heuristic search
- process algebra
- decentralized control