Overcoming Exploration: Deep Reinforcement Learning in Complex Environments from Temporal Logic Specifications.
Mingyu CaiErfan AasiCalin BeltaCristian-Ioan VasilePublished in: CoRR (2022)
Keyphrases
- complex environments
- temporal logic
- reinforcement learning
- concurrent systems
- transition systems
- model checking
- model checker
- reactive systems
- learning agents
- bounded model checking
- formal specification language
- active exploration
- modal logic
- formal specification
- action selection
- autonomous agents
- satisfiability problem
- sequential decision tasks
- agent learns
- formal verification
- belief revision
- mazurkiewicz traces
- linear temporal logic
- state space
- specification language
- computation tree logic
- predicate logic
- temporal knowledge
- multi agent
- verification method
- markov decision processes
- temporally extended
- optimal policy
- learning algorithm
- control flow
- automata theoretic
- temporally extended goals
- dynamic environments
- dynamic constraints
- finite state
- reinforcement learning algorithms
- machine learning