A formal methods approach to interpretable reinforcement learning for robotic planning.
Xiao LiZachary SerlinGuang YangCalin BeltaPublished in: Sci. Robotics (2019)
Keyphrases
- formal methods
- reinforcement learning
- formal analysis
- provably correct
- knowledge based systems
- formal specification
- model checking
- safety critical
- action selection
- planning problems
- model checker
- function approximation
- deterministic domains
- software engineering
- mobile robot
- process algebra
- artificial intelligence
- partially observable
- formal specification language
- learning algorithm
- modeling language
- optimal policy
- real robot
- partially observable markov decision processes
- security properties
- reactive systems
- multi agent
- information systems
- ai planning
- motion planning
- temporal logic
- markov decision processes
- expert systems
- machine learning